European Commission ESPRIT Information Technologies RTD Programme Domain 6: High Performance Computing and Networking Summaries of projects Fourth Framework Programme 1998 Directorate-General III Industry Printed on recycled paper Legal notice Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use which might be made of the information in this document Esprit contact point Esprit Information Desk N-105 8/94 Avenue des Nerviens 105 B-1040 Brussels tel +32 / 2-296-8596 — fax +32 / 2-296-8388 e-mail esprit@dg3.cec.be Web http://www.cordis.lu/esprit/home.html Check for updates to this document on the Web http://www.cordis.lu/esprit/src/projects.htm Legend EU member-states A B D DK E F GR I IRL L NL P S SF UK Austria Belgium Germany Denmark Spain France Greece Italy Ireland Luxembourg Netherlands Portugal Sweden Finland United Kingdom States associated with Esprit CH FL ISL ISR N Switzerland Liechtenstein Iceland Israel Norway Other countries participating in projects CA USA Canada United States of America Roles C P A Coordinator Partner Associate Contractor 20018 PINEAPL Parallel Industrial Numerical Applications and Portable Libraries Keywords: High Performance Computing and Networking Technical - Numerical Applications, Parallel Computing, Portable Libraries, Reliability Uses - Numerical Modelling, Reusable Software, Simulation Industrial Applications - Aerodynamic Analysis, Electromagnetic Analysis, Hydrodynamic Analysis, Thermo-electromagnetic Analysis, Two-stroke Engine Simulation The principal objective of the project is to produce a library of parallel numerical software relevant to a wide range of industries and portable and efficient across a wide range of high-performance machines. The numerical algorithms must be accurate, stable and robust and existing software will be used as much as possible. The end users' application codes have been chosen to represent a varied, if not comprehensive, cross-section of industrial problems including electromagnetics, fluid dynamics, chemical reactions and thermal applications. The numerical software required to solve these problems includes dense, banded and sparse linear algebra, partial differential equations, optimization, FFTs and dynamic load balancing. This choice of application codes will ensure a substantial breadth of applicability for the numerical library. The performance improvements in the end users’applications achieved by the use of the numerical software developed in the project will be a clear metric of its success. Contact Point: M.S. Derakhshan The Numerical Algorithms Group Ltd Wilkinson House, Jordan Hill Road UK - OX2 8DR Oxford tel: +44 1865 511245 fax: +44 1865 310139 email: mishi@nag.co.uk http://www.nag.co.uk/projects/PINEAPL Participants: Country Role The Numerical Algorithms Group British Aerospace Defence IBM SEMEA Math-Tech Thomson-CSF LCR CERFACS CPS The Danish Hydraulic Institute Manchester University Piaggio Veicoli Europei UK UK I DK F F I DK UK I C P P P P A A A A A Start Date Duration January 1996 36 months Octobre 1998 High Performance Computing and Networking / 4 20074 HPC-VAO High Performance Computational Environment for VibroAcoustic Optimization Keywords: High Performance Computing and Networking Technical - Acoustics, Distributed Computing, Parallel Processing, Scheduling, Simulation Uses - Mechanical Engineering, Vibration Analysis, Vibro-acoustic Optimisation Industrial Applications - Automotive and Aerospace Engineering, Home Appliances The project aims to bring together major computer-aided engineering suppliers and automotive manufacturers to develop a high-performance and high-productivity framework for noise and vibration analysis, specifically for vibro-acoustic optimisation. In current practice, several hours of computation on a multi-processor system are needed to analyse a single configuration only. Stricter environmental regulations for external noise and increased customer expectations for both noise and vibration levels are forcing leading automotive companies to improve not only their design methodologies but also the supporting computational environments. The full benefit of vibroacoustic optimisation tools will only be realised if their performance is dramatically increased. This is essential if the product refinement and optimisation phase is to match the demands imposed on development and engineering by competition and stringent time-to-market targets. This project further addresses issues extremely important for improving the productivity of the overall vibroacoustic optimisation process by tackling the major issues of data flow and meshing. Contact Point Jean-Louis Migeot Numerical Integration Technologies NV Ambachtenlaan 11A B-3001 Leuven Participants Country NIT B BMW D IMEC B LMS B MacNeal Schwendler D PAC UK Renault F Rover UK Start Date January 1996 Octobre 1998 tel: +32 16 40 04 22 fax: +32 16 40 04 14 email: jlm@nit.be Role C P P P P P P P Duration 36 months High Performance Computing and Networking / 5 20082 FRONTIER Open System for Collaborative Design Optimisation Using Pareto Frontiers Keywords: High Performance Computing and Networking Technical - Elliptic Parameterisation, Genetic Algorithms, Multiple Criterion Decision Making, Parallel Hill Climbing, Parallel Architectures, Pareto Frontiers Uses - Decision Support, Design Optimisation, Distributed Design, Multidisciplinary Analysis, Industrial Applications - Aircraft, Diesel Engine, Domestic Appliance and Ship Design The project is concerned with multi-partner collaborative design optimisation. This is a key area for future European industrial competitiveness, requiring leading-edge capabilities to improve designs whilst significantly reducing costs and time-to-market. The project will develop a system for product optimisation based on multiple objectives. The most important feature will be flexibility and the ability to accommodate the many diverse techniques currently in use across European industry. Design problems from five major industrial sectors will define the requirements of the system and will validate its effectiveness for diverse industrial applications. The optimisation of a design is highly computationally intensive. The project will use existing HPCN technology and parallel solvers. The system will be developed in three distinct phases, synchronised with user trials in the aerospace, shipbuilding, domestic appliance, heating and diesel engine sectors. Contact Point D C Spicer British Aerospace Defence Ltd Warton Aerodrome UK - PR4 1AX Preston Participants British Aerospace Daimler-Benz Aerospace University of Bergen Universita' degli Studi di Trieste Calortecnica Diesel Ricerche Electrolux-Zanussi Elettrodomestici Kvaerner Ship Equipment University of Newcastle Dera Start Date January 1996 Octobre 1998 tel: +44 1772 855108 fax: +44 1772 856003 email: john.shaw@bae.co.uk Country UK D N I I I I N UK UK Role C P P P A A A A A A Duration 36 months High Performance Computing and Networking / 6 20089 MERCURY Performance Management of Commercial Parallel Database Systems Keywords: High Performance Computing and Networking Technical - Business Work Load, Massively Parallel Platforms, Relational Databases Uses - Data Placement, Sizing Parallel Relational Systems, Tuning, Training, Consultancy Industrial Applications - Performance Management, Supply of Services, System Sizing The objective of the project is to promote the wider use of parallel database systems by helping the user to visualise and understand the way workloads are handled by particular parallel systems and configurations. The project will develop tools to analyse performance and thereby aid the sizing and configuration of commercial systems. These tools will predict the performance of a particular configuration under a specific workload, analyse and display information to highlight bottlenecks and assist the tuning of a particular configuration for a specific workload. The tools to be developed encapsulate knowledge about the database management system and parallel platform to be analysed. These tools will initially support models of the GoldRush, SP2 and nCUBE parallel machines and of the ORACLE and INFORMIX database management systems. A common and easily-used graphical will be developed which will help determine the most efficient overall configuration. This will also provide support for the calibration and validation of the tools which will be done by reference to real commercial workloads giving the project feedback on the applicability of its developments. Results from the project will be used initially by the partners in their daily business. Contact Point Phil Broughton ICL Wenlock Way, West Gorton UK - M12 5DR Manchester Participants ICL IFATEC ING Group CWI Heriot-Watt University Start Date January 1996 Octobre 1998 tel: +44 161 223 1301 fax: +44 161 223 7712 email: pb@wg.icl.co.uk Country UK F NL NL UK Role C P P A A Duration 36 months High Performance Computing and Networking / 7 20111 FSI-SD Fluid Structure Interaction - High-Performance Simulation for Structural Design Keywords: High Performance Computing and Networking Technical: Parallel Computing, Simulation Uses: Computational Fluid Dynamics, Computational Structural Design, Coupled Problems Industrial Applications: Structural, Wind and Offshore Engineering The goal of the project is to enable industry to design fluid-loaded structures more accurately and efficiently by using high-performance modelling and simulation techniques. The consortium aims to address problems such as wind-induced motion of bridges and frame structures and hydrodynamic loads on submerged floating tunnels, pipelines, risers and anchor lines. The work of the project comprises the optimisation of parallel computational fluid dynamics (CFD) codes for fluid-structure interaction (FSI) simulation, the development of adaptive mesh refinement for FSI applications, the modular integrated coupling of CFD and computational structural design modules and the simulation of relevant industrial problems to validate the simulation codes. The project will result in an FSI-simulation code implemented on a parallel computing system. This code will be modular and be open to the interchange of individual components during and after the project. An automatic mesh generator tailored for FSI problems will also be developed. A selected set of case studies from the wind and offshore-engineering sectors will be used for validating the codes. The consortium intends to exploit the results of the project both internally and externally. Contact Point Trond Kvamsdal SINTEF Applied Mathematics N-7034 Trondheim Participants SINTEF FEGS JRC Norsk Hydro Rambøll Start Date January 1996 Octobre 1998 Country N UK I N DK tel: +47 73 59 29 72 fax: +47 73 59 29 71 email: trond.kvamsdal@sima.sintef.no http://tina.sti.jrc.it/FSI/ Role C P P P P Duration 36 months High Performance Computing and Networking / 8 20115 PAROS Parallel Large Scale Automatic Scheduling Keywords: High Performance Computing and Networking, Decision Support, Optimisation Technical - Heuristics, Mathematical Optimisation, Network Computing, Parallelisation Uses - Transport Industrial Applications - Airline Crew Pairing, Rule Compiling The goal of the project is to improve dramatically the speed and quality of automatic scheduling of pairings for Lufthansa. The techniques employed will be kept as generic as possible so as to give them the broadest industrial applicability. Current computer systems are not able to find high quality solutions for very large and complex problems such as pairing construction for major airlines in a reasonable time. Shorter scheduling times will result in less administration and better response times to market requirements. The use of advanced computer solutions will lead to higher productivity and to better flight schedules. The goal of the project will be achieved though the use of parallel computing to give the necessary performance and by refining mathematical optimisation techniques through the improvement modelling techniques and the efficiency of rule management. Contact Point Burkhard Weber Deutsche Lufthansa AG FRA NE 4/V D-60546 Frankfurt tel: +49 69 696 6120 fax: +49 69 696 2662 email: elmuth.eggeling@fra1.lh.lh.dbp.de Participants Lufthansa Carmen Systems Chalmers University of Technology University of Patra Country D S S Role C P P GR P Start Date April 1996 Duration 36 months Octobre 1998 High Performance Computing and Networking / 9 20116 ODESIM Optimum Design of Multibody Systems Keywords: High Performance Computing and Networking Technical - Multibody Design, Multibody Optimisation, Parallel Computing Uses - Computer Aided Design, Simulation, Optimisation of Mechanisms Industrial Applications - Aircraft and Spacecraft, Mechanical Engineering, Vehicles The aim of the project is to develop a complete set of advanced software tools for the optimal kinematic and dynamic design of multi-body systems (MBS) using high-performance, parallel computers. These tools are intended for use in design and industrial production environments relevant to mechanical engineering. The main objectives of the project are: • To develop software modules for kinematic synthesis and optimum dynamic design of MBS using HPCN, linked to a parametric and variational interactive CAD modeller. • An interactive interface for the definition of design variables, geometry and objective functions. • To implement and run the most time-consuming tasks in an HPCN environment making use of multi-CPU computers and clusters of workstations. The starting point for this project will be the CAD and MBS-simulation packages currently available within the consortium. These will be enhanced with additional capabilities that will allow the new modules developed in this project to run in the same environment. The use of standards such as UNIX, C++, X-Windows, Motif and STEP and a modular structure, based on Object Oriented Programming techniques, will guarantee the openness, portability and scalability of the system. The consortium will exploit the software developed both internally and externally. Contact Point José M. Jiménez CEIT Manuel de Lardizabal 15 E-20009 San Sebastian Participants CEIT CASA/Space Division CR Fiat Matra Datavision Siemens CERFACS Start Date January 1996 Octobre 1998 tel: +34 43 21 2800 fax: +34 43 21 3076 email: jmjimenez@ceit.es http://www.ceit.es Country E E I F D F Role C P P P P A Duration 30 months High Performance Computing and Networking / 10 20118 MEMIPS Model Enhanced Solution Methods for Integer Programming Software Keywords: High Performance Computing and Networking, Decision Support, Optimisation Technical - Mixed Integer Linear Optimisation, Mathematical Modelling, Parallel Branch and Bound, Generation of Cutting Planes, Guidance of Solution Process Uses - Chemical Industry, Power Generation Industrial Applications - Distribution, Logistics and Production Planning Many industrial decision processes, production planning problems, distribution problems, and problems occurring in logistics can be described as mixed-integer linear optimisation problems. However, the ability to solve these industrial problems depends on the sophistication of the tools and techniques available. This project will remedy the inadequacies of the tools currently available by developing a flexible, general-purpose commercial software tool for this class of problems. This will provide industrial modellers with access to new, more powerful methods capable of solving problems currently considered intractable. High-performance parallel computers offer the means to obtain solutions to difficult problems within an industrially relevant timeframe. These tools will be developed on parallel high-performance hardware available at all the partners and will be used to solve critical industrial problems at BASF, Unilever, PowerGen and Tractebel. The primary output of the project will be a set of tools which embody expert knowledge and adhere to best practice. The primary results will be much faster solution times for difficult and widerranging problems. This in turn will bring greater operating efficiency, reduced costs, better knowledge and understanding of industrial processes, and improved competitiveness. Contact Point Robert Ashford Dash Associates 50 Binswood Avenue UK - CV32 5TH Royal Leamington Spa tel: +44 1926 315 862 fax: +44 1962 315 854 email: rwa@dash.co.uk Participants Dash BASF Unilever PowerGen Tractebel UCL-CORE University of Buckingham Ziam Country UK D UK UK B B UK D Start Date January 1996 Duration 36 months Octobre 1998 Role C P P A A A A A High Performance Computing and Networking / 11 20124 ESOP A Parallel Optimisation Scheme Applied to Extrusion Die Design Keywords: High Performance Computing and Networking Technical - Computer-aided Design, Computational Fluid Dynamics, Inverse Problems, Parallel Program Development Uses - Medical, Optimisation, Numerical Applications, Precision and Optical Equipment, Simulation Industrial Applications - Chemical and Pharmaceutical Industries, Extrusion, Polymer and Rubber Processing The objectives of the project are to develop an automatic optimisation tool based on Polyflow which takes into account all aspects of die design. In particular this includes flow-balance adjustments, wall-temperature adjustments and front-face design. The project will validate the tool for rubber and PVC profiles and for medical applications. This will allow a comparison with the traditional "trial and error" approach and a report on the economic benefits of the new tool will be a deliverable of the project. The use of the latest HPCN technology will reduce the computing time taken to design a complete die to less than one day on fast parallel computers. Simpler problems will take a similar time on a network of workstations. The problems to be tackled are numerically intensive because they involve non-linear fluid mechanics and optimisation and could not realistically be attempted without the use of high-performance computers. Contact Point Jean-Marie Marchal Polyflow SA Place de l’Université B-1348 Louvain-la-Neuve Participants Polyflow Apex Technologies Hutchinson Kömmerling University College Swansea Vygon Start Date January 1996 Octobre 1998 Country B F F D UK F tel: +32 10 452 861 fax: +32 10 453 009 email: jmm@polyflow.be http://www.polyflow.be Role C A P P P P Duration 36 Months High Performance Computing and Networking / 12 20161 CISPAR Open Interface for Coupling of Industrial Simulation Codes on Parallel Systems Keywords: High Performance Computing and Networking Technical - Coupled problems, Heterogeneous MIMD Architectures, Simulation Uses - Aircraft Design, Automotive Design, Medical Design, Ship Certification Industrial Applications - Crash Simulation, Structural Mechanics The project will develop a general communications library for coupled problems (COCOLIB) and interfaces to it. COCOLIB will enable the coupling of message-passing versions of existing structure and fluid dynamics codes, either developed in-house or provided by independent software vendors (ISVs). Industrial end users in medical engineering, ship certification, automotive and aircraft simulation have identified important coupled problems and have expressed their strong interest to run coupled simulations with the codes PAM-CRASH/PAM-SOLID, PERMAS, and STAR-CD. The corresponding leading European ISVs will cooperate to develop a general coupling interface applicable to these and other codes. Post-processing will allow the simultaneous and synchronised visualisation of representative sets of data from the coupled codes. The objectives of the project are: • To develop a general communications library for coupling codes on parallel systems; • To define open interfaces to be propagated as a standard for coupled problems; • To demonstrate the coupling of leading industrial codes (PAM-CRASH/PAMSOLID, PERMAS, STAR-CD) on parallel platforms; • To extend existing tools to visualise the results of coupled simulations. Contact Point Karl Solchenbach tel: +49 2232 1896 14 PALLAS GmbH fax: +49 2232 1896 29 Hermuelheimer Str. 10 email: solchenbach@pallas.de D-50321 Bruehl http://www.pallas.de Participants Country Role PALLAS D C Computational Dynamics UK P ESI F P Germanischer Lloyd D P GMD D P INTES D P Aerospatiale F A Imperial College UK A Mercedes-Benz D A Sulzer innotec CH A Start Date Duration January 1996 36 months Octobre 1998 High Performance Computing and Networking / 13 20162 PHAROS Open HPF Programming Environment Keywords: High Performance Computing and Networking Technical - High Performance Fortran, Fortran Reengineering, Performance Visualisation, Tools Assessment, Tool Integration Uses - Automotive Design, Defence Industry, Energy Production, Electromagnetic, Fluid and Structural Simulation The project brought together tools for High Performance Fortran (HPF) program development, including the NA Software HPF Mapper and associated toolset, and HPF-oriented support tools from PALLAS and Simulog. These tools stem from prior research and development within ESPRIT and nationally funded projects and have all reached market status. The project aimed to bring these together to form a coherent HPF Toolset which is open and capable of supporting additional and alternative tools and compilers. The project has assessed the effectiveness of HPF and of the current tools with respect to: • The cost of carrying out code parallelisation which should be reduced by using HPF. • The ongoing costs of supporting the parallel code. • The efficiency of the resulting code relative to a hand-parallelised message-passing version. • The effectiveness of the currently available support tools for HPF development and tuning. The assessment has been done on a number of commercial strength codes brought to the project by industrial end users. The codes all suited to the HPF programming style, and for each code a message-passing version already existed against which the comparisons could be made. Contact Point Karl Solchenbach tel: +49 2232 18960 PALLAS GmbH fax: +49 2232 189629 Hermuelheimer Str. 10 email: solchenbach@pallas.de D-50321 Bruehl http://www.vcpc.univie.ac.at/activities/projects/PHAROS.html Participants PALLAS NA Software Simulog VCPC CISE Debis GMD MATRA SEMCAP Start Date January 1996 Octobre 1998 Country D UK F A I D D F F Role C P P P A A A A A Duration 24 months High Performance Computing and Networking / 14 20170 EMCP2 ElectroMagnetic Compatibility using Parallel Parameterisation Keywords: High Performance Computing and Networking Technical - Automatic Optimisation, Electromagnetic Compatibility, Finite Element Solver, Frequency Method, Parallel Computing, Parameterisation, Simulation Uses - Aeronautics Industry, Automotive Industry Industrial applications - Aircraft, Cars and Helicopter Design The project will develop an innovative technique and tool for electromagnetic compatibility (EMC) design using simulation on MIMD computers. A main result of the project will be an optimisation tool allowing interactive simulation on CAD systems used in the initial design. It will allow EMC specialists directly to solve problems and interactively to obtain an optimal design. The technique and tool will be evaluated through the development of different industrial applications from different industrial sectors, namely aircraft, cars and helicopters. The validity of the technique will be established by comparing simulation results with measurements taken either in an anechoic chamber or in an open-air EMC test range. The software developed in the project will allow the European aeronautics and car industries to improve the quality of their EMC designs, to reduce cost of development and time to market of their products and to design safer products. The results of the project will be directly exploited by the partners. Contact Point Marc Viret Aerospatiale Centre Commun de Recherches Louis Bleriot 12 Rue Pasteur F-92152 Suresnes Participants Aerospatiale CCR ALENIA CADOE Centro Ricerche FIAT Ericsson Saab Avionics AB COREP Eurocopter France KTH MIP (CNRS unit) Start Date January 1996 Octobre 1998 tel: +33 1 46 97 37 12 fax: +33 1 46 97 35 00 email: Marc.Viret@siege.aerospatiale.fr Country F I F I S I F S F Role C P P P P A A A A Duration 30 months High Performance Computing and Networking / 15 20183 HOISE-NM HPCN-Europe On-line Information Service Keywords: High Performance Computing and Networking Technical - World Wide Web Use - Electronic News and Information Service for the European HPCN Community This project has created a virtual magazine distributed on the Internet, named Primeur (http://www.hoise.com/primeur). The magazine covers high performance computing and networking (HPCN) specially targeted for industries to promote the use and benefits of these technologies and to share the experiences across a wide range of sectors. Primeur also offers the latest news on the HPCN industries and the centres which use it; it contains information about the programmes and projects of the European Commission and about the key activities on these topics, with links to other WWW servers; and it serves as a forum to exchange ideas and opinions. Primeur is intended as an information source for everyone interested in HPCN in Europe. The articles are short and popular, without technical details but including pointers to get deep when convenient. Readers are not limited to HPCN technicians, but they can include industrial managers and consultants who have not used HPCN technologies yet, public administrators who determine HPCN policies and its usage on other sectors, and anybody interested in HPCN development. Contact Point Ad Emmen Genias Benelux James Stewartstraat 248 NL-1325 JN Almere tel: +31 36 537 3867 fax: +31 36 537 5002 email: emmen@genias.nl http://www.hoise.com Participants Genias CESCA SARA The Eurocommunications group Harms Supercomputing-Consulting HiPerCom NTUA ORAP Polosa Verbion Start Date January 1996 Octobre 1998 Country NL E NL B D UK GR F I SF Role C P P P A A A A A A Duration 24 months High Performance Computing and Networking / 16 20184 HPS-ICE High Performance Simulation of Internal Combustion Engines Keywords: High Performance Computing and Networking Technical - CFD, Combustion, Parallel Program Development, Pre and Postprocessing Uses - Combustion Simulation, Grid Generation, Visualisation Industrial Application - Vehicle Manufacture The overall goal of the project is to integrate the simulation of fluid flow, mixing and heat release in internal combustion engines into the development cycle through the use of HPCN. In order to make further advances in the development of piston engines, including the lowering of harmful emissions and the improvement of fuel economy, it will be necessary to make increasing use of computational fluid dynamics. At present there are two main obstacles to transferring existing relevant basic research to development centres. The first obstacle is the long computational time currently required for the realistic simulation of time-dependent flow fields and turbulent combustion under realengine conditions. The second obstacle is the time and cost required to generate, modify and control the computational grid. The aim of the project is to demonstrate real-engine simulations within one week. These simulations will include the necessary pre- and post-processing and should allow an accurate prediction of velocity fields, turbulence, fuel distribution at ignition point and heat release. This is the necessary condition for achieving the required improvement in the development processes of internal combustion engines from the standpoint of engine designers. From the automotive companies’ viewpoint, the reductions in time, effort and cost to achieve those improvements in engine performance required by market competition and government legislation will represent a major advance and will contribute to their competitive advantage. Contact Point Harald Echtle Daimler Benz AG HPC: E222 D-70548 Stuttgart Participants Daimler-Benz Computational Dynamics Renault RUS Cray Research Start Date January 1996 Octobre 1998 tel: +49 711 17 20884 fax: +49 711 17 52026 email: echtle@str.daimler-benz.com http://www.uni-stuttgart.de/RUSuser/vis/hps-ice.html Country D UK F D D Role C P P P A Duration 36 months High Performance Computing and Networking / 17 20189 PROMENVIR HPC-Based PRObabilistic Mechanical design ENVIRonment Keywords: High Performance Computing and Networking Technical - Distributed Simulation, Finite-element Analysis, Monte-Carlo Simulation, Multibody Analysis, Stochastic Mechanics Uses - Mechanical Parameter Scatter, Stochastic Structural and MBD Simulation Industrial applications - Stochastic FEM and MBD Simulation The project will develop a high-performance tool for performing probabilistic analysis of components such as structures and mechanisms. The underlying technique will be Monte Carlo Simulation which is characterised by intrinsic parallelism and general applicability. A generic platform for a solver-independent approach to stochastic mechanics will be developed whereby either a finite element method (FEM) or multi-body dynamics (MBD) solver may be "plugged-in" without the need for any specific interfacing. The important characteristic of the system will be the treatment of parameter scatter and uncertainty by executing a set of independent deterministic runs generated according to modern sampling schemes which span the entire parameter space. The system will therefore have the capability of running the set of runs on a possibly heterogeneous LAN or WAN, including WANs distributed internationally. This integration of distributed computing facilities will enable the tool to tackle stochastic mechanics and other scientific problems with an unprecedented degree of complexity and detail and with a high industrial relevance and significance. Contact Point V.G. Molinero tel: +34 1 5863796 CASA Space Division fax: +34 1 7474799 Avenida de Aragon, 404 email: vgomez@ana.casa-de.es E-28022 Madrid Participants CASA Space Division CEIT IKOSS ITALDESIGN RUS BLUE Engineering PAC UPC Start Date January 1996 Octobre 1998 Country E E D I D I GB E Role C P P P P A A A Duration 24 months High Performance Computing and Networking / 18 20216 INSIDE Integrated Simulation and Design System for Civil and Structural Engineering Keywords: High Performance Computing and Networking Technical - Adaptive Finite Element Analysis, Computer-Aided Design, Distributed Systems, Error Control, Simulation, Workstation Clusters Uses - Civil and Structural Engineering Industrial Applications - Structural Mechanics, Structural Optimisation and Design The objective of the project is develop a high-performance analysis and design tool for the civil and structural engineering industry based on recent advances in quality-assured finite-element solutions and parallel computing. These will be integrated into a domain-specific computer-aided design system which will meet the rapidly growing demands of the civil and structural-engineering sectors. Amongst these demands are: innovative construction; improved design through better modelling; iterative just-in time design; and levels of safety and economy not achievable with oversimplified models. These demands will be addressed by the speed-up of analysis and modelling through the use of HPCN. An improved quality of analysis which will be achieved by coupling HPCN with advanced finite-element techniques. The prototype system to be developed will be based on existing software, will incorporate the latest research results and will be tested on real-life problems from civil and structural engineering. The project will stimulate the use of HPCN for simulation and design in a sector in which it has previously made little impact. Contact point Casimir Katz Sofistik GmbH Tassiloplatz 7 D-81541 München Participants Sofistik D’Appolonia Prosolvia Sofistik Hellas Technical University of Munich Chalmers University of Technology Heitkamp Technical University of Athens Start Date January 1996 Octobre 1998 tel: +49 89 315878 12 fax: +49 89 315878 23 email: ck@sofistik.mhs.compuserve.com http://www/inf.bauwesen.tumuenchen.de/projekte/inside.html Country D I S GR D Role C P P P P S A D GR A A Duration 30 months High Performance Computing and Networking / 19 20231 TOOLSHED Tools for High Productivity Engineering Design Keywords: High Performance Computing and Networking Technical - Automatic Mesh Generation, Computational Steering, Engineering Design Optimisation, Interactive Environment, Visualisation of Distributed Data Uses - Automotive, Aircraft, Railway, Turbo-machinery Manufacture and Seismic Analysis Industrial applications - Computational Fluid Dynamics, Computational Electromagnetics, Industrial Design Simulations, Structural and Vibration Analysis High-performance computing now allows simulation to be used as a predictive tool in the design of complex components. Such components include combustion chambers, turbo-machinery, civil engineering structures and automotive parts. The design of these involves computational fluid dynamics, computational electromagnetics, seismic analysis and computational structural and vibrational analysis. The objective of the project is to develop a highly productive interactive open environment supporting HPCN simulation techniques applied to engineering problems. The project aims at integrating in a common framework tools developed within other EU projects and elsewhere. The integrated framework will be open to all software packages. This will be achieved by making use of currently available and emerging standards in distributed computing and data. The effectiveness of the environment will be demonstrated in three complex industrial design cycles: • Design of civil aircraft components using computational electromagnetics. • Sensitivity analysis of dams under seismic loads using structural mechanics. • Development of an experimental rig using computational fluid dynamics. Contact Point Gilles Gruez BERTIN & Cie Rue Pierre Curie 59 F-78 373 Plaisir Cedex Participants: BERTIN AEROSPATIALE CISE ENEL GEC ALSTHOM NUMECA PAC RAL SINTEF Start Date February 1996 Octobre 1998 Country F F I I UK B UK UK N tel: +33 1 34 81 88 40 fax: +33 1 30 54 04 14 email: gruez@bertin.fr http://www.bertin.fr Role C P P P P A A A A Duration 30 Months High Performance Computing and Networking / 20 20248 CLOVIS Computing Large Objects Visualisation System Keywords: High Performance Computing and Networking Technical - Data Archiving, Data Compression, Data Fusion, Data Processing, Data Reduction, High Data Throughput Applications, Parallel Architectures, Visualisation Uses - Civil Engineering, Human Modelling, Medical Imaging, Virtual Reality Industrial Application - Industrial Scientific Computing The goal of the project is to develop a portable and efficient tool for the post-processing and realtime visualisation of very large datasets including those too large to output to a network or to store in a raw state in data repositories. Previously it has not been possible to visualise datasets with more than 2563 entities within an acceptable response time. Using data compression, decompression and reduction techniques, The project will make it possible to process and visualise very large datasets within a framework open to existing and emerging standards such as OpenGL, PVM and MPI. The target dataset size to be processed is about 10003 entities taking into account hardware capabilities, memory, I/O bandwidth, CPU and graphics speeds. The project will address currently available multiple-processor architectures including MPP systems, workstation clusters, vector processors, super-servers and high-end workstations, with optimised implementations using standard high performance tools such as MPI, PVM and OpenGL. Using the results of the project, engineers will evaluate their work, exploiting efficiently the results of simulations in a number of areas such as computational fluid dynamics, structural analysis, reservoir simulation, molecular simulation and large-scale discrete and continuous data analysis. The project will address the advanced research, industrial simulation, and educational markets. Contact Point Xavier Morel Tethys 22-24 rue Debertrand F-91410 Dourdan Participants Tethys Intecs Sistemi Tessella Aerospatiale DRA ENEL Air Liquide Adersa Start Date January 1996 Octobre 1998 tel: +33 1 64 59 21 21 fax: +33 1 64 59 21 20 email: x-morel@tethys.fr Country F I UK F UK I F F Role C P P A A A A A Duration 36 months High Performance Computing and Networking / 21 20255 ISIS Interactive Satellite Image Server Keywords: High Performance Computing and Networking, Image Databases Technical - Satellite Image Databases, Thematic Data Access, Wide-area Networks Uses - Earth Observation Information Access and Retrieval System Industrial Application - Agriculture, Coastal Monitoring, Forestry, Geological and Oil Research, Tourism The project aims to develop and test an Earth Observation (EO) information access and retrieval system. The major objective is to allow the demonstration of high-volume, interactive data-access and data-mining services and to offer on-line and thematic data access. The project will address some of the technical constraints to the operational development of EO services. These relate to adequate information extraction tools, data availability and data delivery. In particular, the project will address the improvement of access to products and of their availability and associated delivery times. This will enable users of remote sensing products to have access via computer networks not only to catalogues, but also to the products themselves. The images will be compressed according to individual requirements. To do this the client expresses interactively a priori needs for each requested image. The server selects the appropriate features inside the image, simplifies the information outside those features and compresses the image before sending it to the client. The compressed image is then received and decompressed. The emphasis is on the expansion of the EO market to new users as well as on the improvement of existing capabilities by improving the current level of EO data usage in the supply chain comprising primary data provider, value-added supplier and end user. Contact Point Yves Henaff Matra Cap Systèmes 6 Rue Dewoitine F-78142 Velizy Villacoublay Participants Matra Cap Systèmes Dornier Advanced Computers Systems ITC JRC Spot Image Vitec Multimedia Cap Gemini Cray Systems CSA ESA/ESRIN RS Consult Small End Users Start Date December 1995 Octobre 1998 tel: +33 1 34637272 fax: +33 1 34637320 email: horus@matra-ms2i.fr Country F D I Role C P P NL Int F F I UK IRL Int DK EU Duration 36 months P P P P A A A A A A High Performance Computing and Networking / 22 20259 APEX High Performance Computing for Process Engineering: Awareness and Promotion Exercise Keywords: High Performance Computing and Networking Technical - Computational Fluid Dynamics, Decision Support, Parallel Architectures, Parallel Program Development, Real-time Applications Uses - Multimedia, Process Modelling Industrial Applications - Combustion and Chemical Reaction Simulation The overall objective of the project is to increase awareness in the potential of and to promote the use of High Performance Computing and Networking (HPCN). This will be done by supporting a next-generation approach to industrial R&D in the process industries with emphasis on IT-based simulation and prediction. The aim is to promote widely the results from the HP-PIPES project. This project has developed computational packages and demonstrated that advanced simulation can be used as a realistic predictive tool for the design and operation of reaction and combustion systems. The core code, which models fluid flow and heat transfer in process engineering systems, exploits the power of massively parallel high-performance computing systems. It has been coded in a modular form to facilitate the development of new functions for other applications. The code is already being successfully exploited in a range of collaborative projects involving industry and universities. The end results of the project will be a multimedia demonstration of the application of HPCN and computational fluid dynamics technologies in the accurate prediction of real industrial processes and an interactive package for evaluating the costs and benefits of this class of R&D. There will be two editions of the CD-ROM-based multimedia presentations which will be widely circulated to the process industries and other interested institutions. Contact Point Keith Chessell Paras Ltd 120a High Street Newport, Isle of Wight, UK - PO301TP Participants Country Paras Limited UK CLRC UK IST P Start Date May 1996 Octobre 1998 tel: +44 1983 528700 fax: +44 1983 528800 email:kc@paras.demon.co.uk Role C P P Duration 24 months High Performance Computing and Networking / 23 20956 - RE 1011 - AC 111 JAMES Joint ATM Experiment on European Services Keywords: High-Performance Computing and Networking, Research Networking Technical - Asynchronous Transfer Mode, Multimedia, Trans-European Networks, Quality of Service, Traffic Monitoring, Internet Uses - Technical Testing, Technology Transfer, Telecommunications, Information Society Industrial Applications - Education, Health, Publishing, Telecommunications, Research Collaboration, Information Management This project is financed by the joint research networking initiative of Esprit and the Telematics Applications Programme, and by the ACTS programme. JAMES is a collaborative project between 18 European network operators to research the use of ATM technology for broadband networks. The project aims to develop a Europe-wide ATM-based interconnection structure together with associated experimental services to meet the requirements for implementation of a European High Speed Network for Research in response to a growing user demand for higher bandwidth services. Connectivity for experimental use between the National Research Networks and European research projects (e.g. Esprit projects, National Hosts, ACTS Projects) is provided over an ATM bearer service based on the trans-European network infrastructure established for the European ATM Pilot. The JAMES project is to experiment, test and evaluate the provision of new ATM-based broadband services and applications throughout Europe. These experiments will use traffic originating from National Research Networks and National Hosts. The major objectives of the JAMES project are to: • Establish the user requirements in terms of type of service, features, quality of service metrics, time scale and support, in collaboration with the users. • Prioritise user requirements and define a trial plan for each service required, • Assess the market potential and projections for trans-European broadband services, • Plan, design, build and operate the underlying infrastructure, building on the experience and capabilities of the ATM Pilot, • Validate end-user services and prove end-to-end operability through experimental trials carrying user traffic, • Obtain and exploit feedback from the experimental users (particularly National Research Networks and National Hosts) in order to enable each Network Operator to plan and prepare the launch of acceptable and cost-effective commercial services (Exploitation Plan). A Users Board (UB) has been set up to manage the interface of the project with the users, and provide direct support to users for projects technical problems. Contact Point Pierre Adam tel +33 1 44 44 09 71 France Telecom - Branche Réseaux fax +33 1 44 44 23 75 6 place d’Alleray email: Pierre.Adam@di.france-telecom.fr F - 75505 Paris Cedex 15 http://www.labs.bt.com/profsoc/james/ Octobre 1998 High Performance Computing and Networking / 24 Participants France Telecom BELGACOM BT Deutsche Telekom FINNET International OTE Portugal Telecom Post & Telekom Austria PTT Telecom Netherlands P&T Luxembourg Swiss Telecom PTT Telecom Eireann Telecom Finland Telecom Italia TeleDanmark Telefonica Telenor Telia PT Iceland Bezeq Israel Start Date April 1996 Octobre 1998 Country F B UK D SF GR P A NL L CH IRL SF I DK E N S IS IL Duration 24 months Role C P P P P P P P P P P P P P P P P P P P High Performance Computing and Networking / 25 20957 - RE 1009 TEN-34 Trans-European Interconnect at 34 Mbps Keywords: High-Performance Computing and Networking, Research Networking Technical - Asynchronous Transfer Mode, Multimedia, Trans-European Networks, Quality of Service, Traffic Monitoring, Internet Uses - Technical Testing, Technology Transfer, Telecommunications, Information Society Industrial Applications - Education, Health, Publishing, Telecommunications, Research Collaboration, Information Management This project is jointly financed by the joint research networking initiative of Esprit and the Telematics Applications Programme. The aim is to deploy trans-European networking facilities which will provide complementary international capacity to allow the national research network investments to be exploited on a global basis and maintain the competitiveness of European research. The implementation will address both the requirements for greater capacity and support the emerging multi-media applications. Implementation will be phased, both in terms of geographic and functional roll-out consisting of: • The immediate introduction of a high speed international service based on 34 Mbps access using an ATM Virtual Path based interconnect and the existing Internet Protocol (IP), to those countries which can already exploit a service at this speed, with expansion to other countries as they become ready. The ATM based service will evolve to exploit directly the advanced features that ATM can offer. • In view of the immaturity of ATM technology, the use of an ATM test bed is also required to validate new developments and features in advance of their introduction in pilot services. Collaboration with the JAMES project has been agreed to use their experimental ATM network for this purpose. In order to maintain connectivity between all European research networking facilities, a continuing interconnection between the new high speednetwork and the existing network services via gateway services will be provided. Contact Point Cathrin Stoves tel +44 1223 302992 DANTE fax +44 1223 303005 Lockton House, Clarendon Road email: Cathrin.Stoves@dante.org.uk UK - CB2 2BH Cambridge http://www.dante.net/ten-34.html Participants Country Role DANTE UK C ACOnet A P DFN D P FCCN P P GSRT GR P INFN I P NORDUnet SF,N,DK,S P RedIRIS ES P RENATER F P RESTENA L P SURFnet NL P SWITCH CH P UKERNA UK P OSTC B P HUNGARNET HU P CESNET CZ P ARNES SI P Start Date : February 1996 Duration : 30 months Octobre 1998 High Performance Computing and Networking / 26 20966 MICA A Model for Industrial CFD Applications Keywords: High Performance Computing and Networking Technical - Computational Fluid Dynamics, Internet, Simulation, Virtual Reality Uses - Fires in Buildings, Furnace Simulation, Urban Pollution, Building Simulation Industrial Applications - Furnace Design, Building Design, Environmental Modelling. The objective of the project is to develop an infrastructure, MICANET, whereby industrial end users will have access the power of computational fluid dynamics (CFD) software running on highperformance computers at a reasonable cost. The elements of MICANET are customised virtual reality interfaces, internet for transfer of data and centres with powerful computers, CFD-simulation software and expertise. Ten initial application sectors have been selected to demonstrate the MICANET capabilities. These have been divided into two groups from industry and construction. Applications from the former group include coal-fired, metallurgical and glass-melting furnaces and industrial oven and heat exchangers. Those from the latter group include urban pollution, flow around buildings, heating, ventilation, air conditioning, fires in buildings and oil-platform explosions. As the project progresses, and after its completion, it is intended that the range of application sectors will broaden and that other simulation centres will be incorporated to increase the value of the MICANET concept. Contact Point John Heritage tel: +44 181 947 7651 CHAM Ltd fax: +44 181 879 3497 Bakery House, 40 High Street email: jrh@cham.demon.uk http://www.cham.co.uk/mica.html. UK - SW19 5AU Wimbledon Participants Country Role CHAM Ltd UK C INRIA F P LSTM-Erlangen D P NTUA GR P University of Zaragoza E P Building Research Establishment UK A Christian Michelson Research N A Hoogovens Groep NL A Paderborn Centre for Parallel Computing D A Stork-Comprimo NL A Swedish Meteorological and Hydrological Institute S A Technical University of Lisbon P A Vattenfall Utveckling S A Wimberly Allison Tong & Goo UK A Start Date Duration January 1996 28 months Octobre 1998 High Performance Computing and Networking / 27 21007 IOTA Intelligent Object Tracking Using Smart Sensor Technology Keywords: High Performance Computing and Networking Technical - Embedded Systems, Digital Signal Processors, Image Processing, Large Data Throughput, Optical Processors, Real-time, Smart Sensors Uses - Machine Vision, Medical Diagnosis, Virtual Reality, Vision Systems Industrial Applications - Medical Imaging, Object Tracking, Optical Sensors, Robotics The objective of the project is to develop highly programmable imaging systems for a wide range of industrial applications which can be tightly integrated with dedicated processors and sensors. To demonstrate the system an intelligent object tracking and analysis system will be implemented addressing the following applications: the integration of a high-resolution sensor and parallel processing array for the real-time, on-line measurement of three-dimensional eye movements; and the integration of a dedicated sensor and processor configuration for contact-free measurement of the point-of-regard for the active control of computer systems and other equipment by visual fixation. Central to the project is the exploitation of present state-of-the-art VLSI technology to develop a high-resolution smart-sensor chip. Using current sub-micron CMOS processes will yield a chip design capable of supporting a pixel density of more than 5122. Based on existing smartsensor technology developed by Integrated Vision Products this device would permit sampling rates of up to 1000 images per second. A control interface will be developed for the programming of the imaging elements and the on-chip processors. A high-speed data interface to a scalable array of signal processors will also be developed. This configuration will permit a wide range of applicationspecific algorithms to be implemented for the identification and analysis of complex objects. A further general-purpose interface will communicate with standard computer systems. Contact Point E. Schmidt tel: +49 3328 430 300 SensoMotoric Instruments GmbH fax: +49 3328 430 320 Potsdamerstraße 18a D-14513 Teltow Participants Country Role SensoMotoric Instruments D C Freie Universitat Berlin D P Integrated Vision Products S P EPFL CH A Universidad Jaume I de Castello E A Start Date Duration June 1996 36 months Octobre 1998 High Performance Computing and Networking / 28 21012 EFTOS Embedded Fault Tolerant Supercomputing Keywords: High Performance Computing and Networking Technical - Embedded Systems, Fault Tolerance, MIMD Architectures Uses - Parallel Program Development, Programming Tools Industrial Applications - Industrial Process Control, Real-time Applications, Embedded Systems The main objective of the project is to provide a software framework for the implementation of fault tolerance in embedded supercomputing applications. This framework will provide a flexible, standard approach for programmers making embedded supercomputing applications more dependable. This approach will shorten the time to market, facilitate software development, maintenance and upgrading, and reduce dependence on specific hardware. Fault-tolerance functions will be accessible to application developers through a high-level portable software library or through ParsyFrame, the high-level API for developers of applications on Parsytec systems. This will hide hardware-dependent features from the application programmer, guaranteeing that the application can be ported to standard hardware platforms and real-time kernels.The target architecture of the project is an MIMD, homogeneous, distributed-memory, message-passing, multi-processor system. Work will be performed on the Parsytec CCe (PowerPC/Parix), the new Parsytec CC/ipp system (Intel/NT) and on DECs Alpha based boards for embedded systems. Important industrial end-user applications of embedded parallel computing will drive the requirements and demonstrate the results. The applications include mail sorting (Electrocom) and control of high-voltage substations (ENEL). Contact Point Gerhard Peise Consultant Wilhelm-Ziemons-Str. 81 D-52078 Aachen tel: +49 241 9 28 58-0 fax: +49 241 9 28 58-99 email: gerhard_peise@compuserve.com Participants Parsytec Computer Electrocom ENEL TXT DLR KU Leuven NTUA Country D D I I D B GR Start Date April 1996 Duration 24 Months Octobre 1998 Role C P P P A A A High Performance Computing and Networking / 29 21017 INNOVA High Performance Telecontrol Station with Image and Sound Capabilities Keywords: High Performance Computing and Networking Technical - Embedded Systems, Databases, Data Fusion, Image Processing, Multi-sensor Systems, Pattern Recognition, Real-time Systems, Sound Processing Uses - Monitoring, Process Control, Real-time Control and Other Applications, Remote Process Monitoring, Remote Terminal Units Industrial Applications - Automobile, Energy, Pharmaceutical and Steel Production The project aims to develop software to be embedded in the new generation of high-performance computing tele-control stations to be integrated in supervisory control and data acquisition systems. These stations will deal not only with digital and analogue signals produced by conventional process sensors, but also with images and sounds. With their ability to obtain process information through a real-time fusion of conventional control signals, images and sounds, these stations will allow the development of more complete, flexible, reliable and secure control systems capable of solving difficult industrial control problems. To perform all the necessary functions and still satisfy the requirement for real-time operation, the software will have to run on a high-performance computer. For exploitation reasons the system to be developed must be low-cost and highly configurable. A steel-manufacturing process will serve to demonstrate the operation of the system in a real-world pilot application. Such an industrial process has been selected because it has specific control problems that are not well solved by conventional systems, and because of the extreme operating conditions present in steelworks. The successful implementation of the system will allow the end-user to obtain qualitative and quantitative improvements in the operation of the steelmaking process. Contact Point Jesus Tardon ELIOP SA Avda. de Manoteras s/n E-28050 Madrid Participants ELIOP EID ESTEBAN ORBEGOZO UNINOVA Start Date May 1996 Octobre 1998 tel: +34 1 302 39 40 fax: +34 1 302 92 49 email: tardon@eliop.es Country E P E P Role C P P P Duration 30 months High Performance Computing and Networking / 30 21023 CATIE Colour and Texture Inspection Equipment Keywords: High Performance Computing and Networking Technical - Embedded Systems, Pattern Recognition, Heterogeneous Architecture, Real-Time Applications, Sensor Modelling, Scalability, Inspection Systems Uses - Embedded Systems Design, Parallel Program Development, Quality Control Industrial Applications - Steel Industry, Timber Industry, Food Sorting The objective of the project is to provide cost-effective colour and texture-based automatic inspection and sorting solutions for industry. Three application areas will be considered: hot steel strip, timber and food. Novel solutions for low-delay image analysis, allowing a few tens of milliseconds from imaging to decision, will be developed and used in real-time on-line inspection demonstrators for each application area. The platform will cope with the high volumes of data associated with colour and texture inspection. The basic technology underlying high-performance personal computers will be used. Because of the vibrations of hot steel strip and the nature of the motion of food particles to be sorted, high-speed prism-based colour line-scan cameras will be developed to capture the RGB values of each pixel simultaneously. Due to the required high line-scan rates, stable illuminators with feature-enhancing radiation patterns will be designed, and an on-line colour camera calibration technique will be developed to make the colour measurements independent of longer term changes in illumination. The consortium intends to exploit the results of the project by marketing and selling inspection systems to industrial end-users, system platforms and dedicated hardware modules to system integrators and by using the developed inspection systems in their own production. Contact Point Timo Piironen Spectra-Physics VisionTech Oy Teknologiantie 2 P.O. Box 80 SF-90571 Oulu Participants Spectra-Physics VisionTech ELEXSO Sortiertechnik Junckers Industrier STN Atlas Elektronik T.V.I. - Temet Vision Industry Fraunhofer Gesellschaft IITB Technical University of Denmark University of Oulu VTT Electronics Start Date January 1996 Octobre 1998 tel: +358 81 551 4291 fax: +358 81 551 4556 email: Timo.Piironen@spvs.elisa.fi http://tks.oulu.fi/Catie/ Country SF D DK D SF D DK SF SF Role C P P P P A A A A Duration 36 months High Performance Computing and Networking / 31 21028 PACHA Parallel Scalable Computers for High Performance Applications Keywords: High Performance Computing and Networking Technical - CORBA, OS Kernel, Parallel Processing, PCI, Real-time, SCI Uses - Control, Databases, Multimedia, Simulation Industrial Applications - Electromagnetic Measurement, Pharmaceutical Database Applications The project intends to develop a high-performance parallel processing system and associated tools allowing to develop enterprise systems as applications based on standard components. The consortium will port four applications in the industrial, scientific and pharmaceutical domains. Emerging standards both in software and hardware will be exploited to investigate the design and implementation of fully featured CORBA-compliant software based on SCI interconnection technology. The work will investigate the scalability of systems using a combination of high performance CPU and SCI, providing high data transfer rate, low latency and cache coherence. The goal is to define a software approach that produces standard based portable and scalable software applicable to the above hardware configurations. It is intended to investigate and exploit hardware platforms scalable from a few to several hundred processors. It is expected to demonstrate the system in the following applications: concurrent engineering, simulation, control systems, highperformance data warehouses and molecular databases. Contact Point Chi Ngo Duc Spacebel Informatique SA I. Vandammestraat 5-7 B-1560 Hoeilaart Participants Spacebel Aérospatiale Dolphin Fraunhofer Intecs Sistemi Telmat University of Basel Novartis INRIA IMD Start Date February 1997 tel: +32 2 658 20 65 fax: +32 2 658 20 90 email: chi.ngo_duc@spacebel.be http://www.spacebel.be/projects/pacha.html Country B F N D I F CH CH F I Role C P P P P P P A A A Duration 36 months This page is located at www.cordis.lu/esprit/src/21028.htm It was last updated on 18 November 1998, and is maintained by Massimo.Luciolli@cec.be Octobre 1998 High Performance Computing and Networking / 32 21036 EIVIS Embedded Interactive Video Server Keywords: High Performance Computing and Networking Technical - Embedded Systems, Distributed Memory, Fault Tolerance, MPEG, Multimedia Systems, Real-time systems, Video-on-Demand Uses - Entertainment Systems, In-flight Entertainment Systems Industrial Applications - Aircraft, Ferry, Hotel and Train Video Systems The objective of the project is to prepare for the development of future Video-On-Demand (VOD) products for the transportation market. It is envisaged that, in the near future, commercial planes, ocean liners, ferries and trains will be fitted with VOD servers. Passengers will then be able to choose and control video programmes from individual terminals. The server will also provide games, tourist information, news and other services. A strong growth in this type of services and products is anticipated in the near future if such systems can be made available with an acceptable performance at a reasonable price. The project will evaluate the suitability of HPCN technology for an embedded VOD server. This will be supported by the development of a demonstrator including a server with an operating environment and applications. Evaluation of this demonstrator will require the presence of a large number of terminals, a few of which will be physical while the others will be simulated. This will enable the project partners to implement a pilot VOD application and to debug, test and optimise the whole system from a functional and a technical point of view. The system will be largely based on standard components and interfaces. Specially designed software tools will be used for system tuning. Contact Point Francis Atche Dassault Electronique 55, quai Marcel Dassault F-92214 Saint Cloud Participants Dassault Electronique Becker Flugfunkwerk GMD Hyperparallel Technologies Sistemas Expertos Iberia Lineas Aéreas de España RENFE Start Date January 1996 Octobre 1998 Tel: +33 1 34 81 67 89 Fax: +33 1 34 81 48 51 Email: francis.atche@dassault-elec.fr Home Page Country F D D F E E E Role C P P P P A A Duration 26 months High Performance Computing and Networking / 33 21037 PCI-II Parallel Computing Initiative II Keywords: High Performance Computing and Networking Technical - Parallel Architectures, Parallel Platforms, Parallel Program Development, Porting The initiative aims to promote HPC technology and culture within European industries by encouraging the application of parallel systems in particular. The project is organised as a cluster of ten subprojects managed overall by CEPBA-UPC. The global activities include project management, monitoring of the work, common services and dissemination of results. Each subproject ports or develops an application for a parallel platform. With some minor exceptions the work within a subproject is structured in three major phases: design and parallelisation approach; prototype; and final system. Each of the projects will develop a system or tool that will improve the industrial competitiveness of the end users. The impact of the individual subprojects will be increased through global dissemination activities across the whole cluster. Industrial partners within each subproject are committed to promote and exploit the tangible developments of the subproject in their respective markets. Experience gained by the universities and research centres will also be exploited within the scientific community. The global activities are directed towards increasing the impact of HPCN in industry. The experience and results gained from the project will be used to continue the promotion of parallel processing and encourage its wider uptake by industry. Contact Point Jesus Labarta CEPBA-UPC C/Gran Capitan, s/n Mod. D6 E-08071 Barcelona Participants Country CEPBA-UPC E Start Date April 1996 Octobre 1998 tel: +34 3 4016987 fax: +34 3 4017055 email: jesus@ac.upc.es http://www.ac.upc.es/cepba/ Role C Duration 18 months High Performance Computing and Networking / 34 21037 PCI-II AMODES Anti Money Laundering Decision Support System Keywords: High Performance Computing and Networking Technical - Data Management, Knowledge-based System, Parallel Architectures, Parallel Databases Uses - Bank Management, Money-laundering Countermeasures , Information Management The main objective of the project is the evaluation of a combination of advanced information technologies, namely data analysis, data management and high-performance computing, in the context of a data warehouse within one of the most technologically mature Italian financial institutions. The data warehouse will be designed with the practical objective of providing support for money laundering countermeasures. The project will develop software modules supporting both data access and data management. These impact of these modules will be characterised particularly with respect to performance aspects. The project will cover many of the central aspects involved in the design of a data warehouse specifically its architecture, its relation to operational systems, performance issues, the development of end-user applications and methods of access. Contact Point Paolo Stofella AIS SpA Via Esterle, 9 I-20132 Milano Participants AIS CariVerona Catholic University PAC Start Date April 1996 Octobre 1998 tel: +39 2280141 fax: +39 2810853 email: paolos@ais.it Country I I I UK Role C P P P Duration 18 months High Performance Computing and Networking / 35 21037 PCI-II GLASSPAR Parallelization of a Glass Manufacturing Modelling Code Keywords: High Performance Computing and Networking Technical - Finite Element, Modelling, Parallel Architecture, Pre and Post-processing, Simulation Uses - Modelling Glass Manufacturing Industrial Applications - Glass-manufacturing Process Simulation, Vat Design The objective of the project is to provide the end user, Cristaleria Española, with an interactive tool to design the geometry of the tin vat in which glass is produced. This is important to improve quality and to reduce manufacturing costs. The tool will be a parallel version of an existing hydrodynamic and thermal code. The sequential version of the modelling code exhibits large compute times, greater than seven hours, even when it runs on a powerful modern workstation. Present practice is to produce thinner and thinner glass. Any irregularities impair the reduction in thickness. For thin glass the number of sheets passing quality control tests is consequently low. It is of considerable economic importance to improve the design of production vats which would enhance the quality of glass produced. A significant improvement in quality would reduce manufacturing costs and substantially improve the competitiveness of companies achieving that improvement. Contact Point Raman Rodriguez Cristaleria Española Avenida de Lugo, 110 E-33400 Aviles Participants Cristaleria Española CEPBA-UPC Universidad de Cantabria Start Date April 1996 Octobre 1998 tel: +34 85 127151 fax: +34 85 127137 Country E E E Role C P P Duration 18 months High Performance Computing and Networking / 36 21037 PCI-II HIPEROAD High Performance Road-vehicle Optimized Aerodynamic Design Keywords: High Performance Computing and Networking Technical - CFD, Optimisation, Parallel Architectures, Pre and Post-processing, Workstation Clusters Uses - Aerodynamic Design, Automatic Optimisation Systems Industrial Application - High-performance cars The project aims to develop a parallel software package to speed up significantly the preliminary design of a high-performance road vehicle. The software will maximise a chosen aerodynamic merit function based on shape whilst satisfying geometrical and operational constraints from the stylist and the manufacturer. An automatic optimisation procedure will be used, based on the repeated evaluation of the aerodynamic forces acting on the vehicle. A routine, whose effectiveness has already been demonstrated, solves the "attached flow" equations by means of the panel method and by modelling the wake of the car. This permits a good evaluation of the vertical aerodynamic loading on carefully shaped high-performance cars, even if the drag cannot be accurately predicted. The code will be written in Fortran and C using PVM or MPI to support communication. The parallel programming environment, TRAPPER, will be used to support code development. The developments will target workstation clusters available to each of the consortium partners. Exploitation will occur both within the consortium and externally through Genias. Contact Point Roberto Fedeli Ferrari Auto SpA Via Abetone Inferiore, 4 I-41053 Maranello (Modena) Participants Ferrari Genias P3C Start Date April 1996 Octobre 1998 tel: +39 536949405 fax: +39 536943057 email: gvd@ferrari.it Country I D I Role C P P Duration 18 months High Performance Computing and Networking / 37 21037 PCI-II MMIPPS Multispectral and Multitemporal Image Processing on Parallel Systems Keywords: High Performance Computing and Networking Technical - Distributed Memory, Image Processing, Parallel Architectures, Real-time processing Uses - Multi-spectral and Multi-temporal Image Processing, Remote Sensing Industrial Applications - Cartography, GIS The main objectives of the project are: • To identify image processing functions required by end users which can benefit from HPCN; • To develop an open portable image-processing software package with the required functionality. The package will be based on commercial off-the-shelf hardware and software components in order to reduce system costs and to limit the developments neeeded. The main outputs of the project are: • A fully operational multi-spectral and multi-temporal image processing parallel software package running on appropriate hardware; • A generic architecture for the software package with adequate documentation and installation instructions for a wide range of platforms. The results of the project will be exploited directly through the consortium partners, both internally and externally. Contact Point Salvo Sabina Intecs Sistemi SpA Via L Gereschi, 32/34 I-56127 Pisa Participants Intecs Sistemi Geospace Gepard Italeco University of Leiden Start Date April 1996 Octobre 1998 tel: +39 50545232 fax: +39 50545200 email: salvo@pisa.intecs.it Country I A A A NL Role C P P P P Duration 18 months High Performance Computing and Networking / 38 21037 PCI-II PARSAI Parallel Simulation in the Automotive Industry Keywords: High Performance Computing and Networking Technical - CFD, Domain Decomposition, Parallel Architectures, Pre and Post-processing Uses - Automotive and Manufacturing Simulation Industrial Applications - Automotive Design, In-car Flow Simulation, Cylinder Simulation The main objective of the project is to demonstrate the applicability of parallel high-performance computing to manufacturing industry through the use of the parallel VECTIS flow simulation code to solve problems specified by end users from the manufacturing sector. The project will simulate the full operational cycle of a cylinder and port system in less than two days. This represents a reduction in computational time of one order of magnitude compared with the sequential code. Such as rate of simulation will enable the design of experiments to be CAE-led and may reduce the overall development cycle by a factor of up to 50. The parallel VECTIS code is a prototype which will be enhanced incorporating those new features necessary for the cylinder simulations such as fuel-spray modelling and through its integration into the sequential environment. The resulting code will be evaluated by the end user and an assessment of the business benefits of the parallel code will be made. The code will be portable across shared and distributed-memory parallel platforms. The code will be exploited both internally and externally by Ricardo. Contact Point Mike Surridge PAC Venture Road, 2 UK - SO16 7NP Southampton Participants PAC Ricardo Volkswagen Start Date April 1996 Octobre 1998 Country UK UK D tel: +44 1703 760834 fax: +44 1703 760833 email: ms@pac.soton.ac.uk http://www.pac.soton.ac.uk/ Role C P P Duration 18 months High Performance Computing and Networking / 39 21037 PCI-II PARSAR Parallelization of the Chirp Scaling Algorithm SAR Processor Keywords: High Performance Computing and Networking Technical - FFT, Image Processing, Parallel Architectures, PVM, Workstation Clusters Uses - Cartography, Earth Observation, Remote Sensing Industrial Application - SAR Images The objective of the project is to parallelise the chirp scaling algorithm (CSA) for use in synthetic aperture radar (SAR) applications. The CSA is a new SAR processing method which has important advantages over other techniques in terms of both accuracy and efficiency. An exploitation plan has been defined and the following applications of the parallel CSA SAR processor are envisaged: • Production of SAR images for remote sensing companies following specific user requirements; • Large-scale production of SAR images at ground stations. processing and archiving facilities and earth observation centres; • Near real-time SAR applications such as the detection of ships and oil spills; • The mass production of digital terrain models using SAR interferometry. Contact Point Antonio Martinez tel: +34 1 3963935 Indra-Espacio fax: +34 1 3963912 C/Mar Egeo s/n email: amar@mdr.inisel-espacio.es E-28830 San Fernando de Henares Madrid Participants Country Role Indra-Espacio E C CEPBA-UPC E P Infocarta E P TSC-UPC E P Start Date Duration April 1996 18 months Octobre 1998 High Performance Computing and Networking / 40 21037 PCI-II PCECOWATER Parallel Computing of Environmental Coastal and Lake Shallow Water Dynamics Keywords: High Performance Computing and Networking Technical - Finite Element Modelling, Parallel Architectures, Simulation Uses- Fluid Modelling, Water Circulation Industrial Applications - Environmental and Hydraulic Engineering, Power Plant Studies The project aims to develop parallel software for studying the circulation of shallow water. This will promote parallel computing in environmental modelling, specifically applied to shallow rivers, lakes and coastal regions. The consortium will develop the parallel code starting from serial versions of finite element modelling software from the SWEET and TELEMAC packages. The results of the project will comprise: • A reduction in the computation time to perform parametric analyses for optimised plant design; • A reduction by an order of magnitude of the computation time for environmental modelling; • The capability to simulate large areas up to 50 Km2 with a detailed representation of regions of a small as 100 m2. This is not currently possible on workstations. The parallel software will be used by the industrial partners for internal and external studies related to the construction and exploitation of power plants and to the TELEMAC customer- base which currently stands at around 20 sites. Contact Point Paolo Molinaro Enel-Cris Via Ornato, 90/14 I-20162 Milano Participants Enel-Cris CRS4 EDF-LNH Kemijoki Politecnico di Milano Reiter Start Date April 1996 Octobre 1998 tel: +39 272248463 fax: +39 272248540 email: molinaro@cris.enel.it Country I I F SF I SF Role C P P P P P Duration 24 months High Performance Computing and Networking / 41 21037 PCI-II RAPT-2 Radiotherapy Application Deployment on Parallel Technology Keywords: High Performance Computing and Networking Technical - Monte-Carlo Simulation, Parallel Architectures Industrial Applications - Medical Diagnosis, Therapy, Treatment Planning The project will build on the achievements of the EUROPORT RAPT project by addressing practical deployment issues. A demonstration system comprising both hardware and software will be installed in a hospital radiotherapy department. Initially a retrospective medical trial will be conducted using archived patient data looking to improve on treatment plans determined by conventional means. This will both validate the simulation method and verify the medical benefits to patients from the RAPT approach. The results will be exploited in two ways. The end-user hospitals will use the work of the project as the basis for further research into improved methods for radiotherapy and for other diagnostic or therapeutic applications of Monte-Carlo simulations. The system will be exploited commercially in partnership with vendors of associated medical equipment such as CT scanners and radiotherapy beam equipment. It is estimated that around 500 radiotherapy systems are needed in the EU and the associated treatment planning has a cost of around 100 million ECU per year. Contact point Mike Surridge PAC Venture Road, 2 UK-SO16 7NP Southampton Participants PAC EDS Le Molinette (Torino) Ospedali Galiere Parsytec Start Date April 1996 Octobre 1998 Country UK I I I D tel: +44 1703 760834 fax: +44 1703 760833 email: ms@pac.soton.ac.uk http://www.pac.soton.ac.uk/ Role C P P P P Duration 12 months High Performance Computing and Networking / 42 21037 PCI-II RETACO Reactive Transport Codes of Use in Environment Keywords: High Performance Computing and Networking Technical - Finite Element, Parallel Architectures, Reactive Transport Model Use - Aquifer Simulation Industrial Applications - Aquifer Pollution Studies, Environmental Consulting The main objective of the project is to develop a parallel code to simulate the transport and decay of reactive solutes in aquifers. This will be achieved by parallelising the EXCHANGE code developed at DIT-UPC. The resulting code, which will need to exploit parallel hardware to have a high enough performance, will have a unique market position. It should become a basic tool for studying aquifer pollution problems and for devising remedial strategies. The development of the parallel code will allow QuantiSci to apply more complex reactive transport models with benefit to their software package, SPADE. This will enable more complex metal decontamination problems to be tackled and for more certain and quantifiable remedial actions to be proposed. Contact Point Jordi Bruno QuantiSci SL Parc Tecnologic del Valles E-08290 Cerdanyola - Barcelona Participants Country QuantiSci E CEPBA-UPC E DIT-UPC E QuantiSci UK Start Date April 1996 Octobre 1998 tel: +34 3 5824410 fax: +34 3 5824411 email: jbruno@quantisci.es http://www.quantisci.es Role C P P P Duration 18 months High Performance Computing and Networking / 43 21037 PCI-II STAMPAR Development of a Proposal Software for Enhanced Design of Sheet Stamping Dies Keywords: High Performance Computing and Networking Use - Design of Sheet Stamping Industrial Application - Sheet Stamping The project aims to parallelise the serial sheet stamping code STAMPACK. The code is based on an explicit finite-element formulation and deals with large strains, elastoplastic materials and frictional contact conditions. STAMPACK was developed by CIMNE in a BRITE project for the numerical simulation of sheet stamping. It is expected that the enhanced parallel version of STAMPACK will reduce computational times by one order of magnitude for solving practical problems in the die manufacturing industry. The code will be implemented using a suitable message-passing system such as PVM or MPI and will run initially on a network of PCs available at the end user Candemat. The software will be easily portable to a network of workstations. Candemat plan to use the software for production as soon as it becomes available. The code will be exploited externally by CIMNE supported by CEPBA. Contact Point Eugenio Oñate CIMNE C/Gran Capitan s/n Mod. C1 E-08071 Barcelona Participants CIMNE Candemat CEPBA-UPC Start Date April 1996 Octobre 1998 tel: +34 3 2057016 fax: +34 3 4016517 email: cimne@etseccpb.upc.es Country E E E Role C P P Duration 18 months High Performance Computing and Networking / 44 21040 EUROPRO European Processing Core Technology Keywords: High Performance Computing and Networking Technical - Embedded Systems, MIMD Architectures, Simulation Uses - Video image Processing, Sound Imaging The objective of the project is to develop a new generation of HPCN embedded platforms and a corresponding software development environment. These platforms will feature a competitive cost/performance ratio and cost. The multiprocessor software development environment will allow a comprehensive mix of computational models comprising a complete set of modelling and simulation tools. The architecture of the hardware will integrate SHARC and PowerPC processors, Direct Switching Technology from Thomson Airsys and link technology from Bull. The software technology will come from Intecs Sistemi and Simulog who will provide modelling, simulation and monitoring tools. Exploitation will come through the partners both internally for their own business uses and externally through Cetia. In particular Thomson Communications and Dicesva will use the system in their next generations of video image processing and sound imaging systems. Contact Point Dominique Devaux Thomson Marconi Sonar SAS 525 Route des Dolines BP 157 F-06903 Sophia Antipolis Participants Thomson Marconi Sonar Bull Cetia Dicesva GEC Marconi Thomson CSFCommunications Fraunhofer ICR Intecs Sistemi LGAI Simulog Thomson CSF-Airsys UPMC Start Date January 1996 Octobre 1998 tel: +33 4 92 96 46 85 fax: +33 4 92 96 39 84 email: dominique.devaux@tms.thomson.fr http://www.cetia.com/news.EUROPRO Country F F F E UK F Role C P P P P P D E I E F F F A A A A A A A Duration 36 months High Performance Computing and Networking / 45 21042 CTIAC Computational Tools and Industrial Applications of Complexity Keywords: High Performance Computing and Networking Technical - Design and Control of Complex Systems, Embedded Systems, Neural Networks, Pattern Recognition, Prediction Uses - Cardiogram Analysis, Optical Coherence Tomography, Quantum Nanoelectronic Systems, Remote Sensing, Unstable Complex Systems Industrial Applications - Agriculture, Aircraft and Car Manufacture, Cardiology, Dermatology, Management of Information Systems and Plants, Microelectronics The objective of the project is to develop new powerful and robust analytical and numerical algorithms for the analysis, prediction and control of complex systems of different types and to produce relevant software for industrial applications. Complexity, self-organisation, chaos and resonances are key features of all realistic systems. These features demand new computational tools for information processing in unstable systems. The project will demonstrate the practical use of these new tools to complex systems by addressing seven industrial applications. The envisaged applications include advanced analysis of electrocardiograms, in-vivo imaging of human skin tissues for the diagnosis of burns and skin cancer, new quantum nanoelectronic devices, simulation of turbulent flow, control methods for the management of information systems and plants, and use of remote sensing images for agriculture and environmental monitoring. The consortium will exploit the results of the project by licensing the software to industrial end users and by the development of an optical coherence tomography product which will meet dermatological requirements for the treatment of burns, tumours and melanoma and market needs for estimating the diffusion of cosmetics within the skin. The project is a joint collaboration among the participants listed below together with a network of collaborators from the International Solvay Institutes from Europe and Russia. Contact Point Ioannis Antoniou InternationalSolvay Institutes Campus Plaine ULB CP231 Bd. du Triomphe B-1050 Brussels Participants International Institutes Lynton Lasers Parsytec Computer Siemens AG Start Date April 1996 Octobre 1998 tel: +32.2.650.55.53 fax: +32.2.650.50.28 email: antoniou@solvayins.ulb.ac.be http://solvayins.ulb.ac.be/ Country Solvay B UK D D Role C P P P Duration 36 months High Performance Computing and Networking / 46 21050 SHAROC System for High Speed Automatic Reading Of Cheques Keywords: High Performance Computing and Networking Technical - Embedded systems, Pattern Recognition, Data Fusion, Image Processing Use - Optical Character Recognition Industrial applications - Mail Order Companies, Banks, Data Processing and Related Services, Finance and Insurance, Outsourcing companies, Software Consultancy and Supply The first part of the project, completed in August 1997, led to the definition and prototyping of a new high-speed automatic cheque processing ystem focused on the UK market. Thanks to cooperative recognition of both legal and courtesy amounts, more than 50% of cheques were individually read with less than 1% error. Moreover, if fused with another external courtesy amount reader, this performance increased up to 70% read rate with still less than 1% error rate. A large-size mail order company, Empire Stores, is participating in the second phase of the project as an end-user to validate this technology in real operation and to improve system performance thanks to on-site experience. It is planned to later extend the SHAROC system to the processing of other documents such as forms or invoices, in order to propose customers an integrated solution taking into account all documents they have to process in large volumes. Contact Point Brigitte Plessis tel +33 1 34 63 72 37 MATRA Systèmes et Fax +33 1 34 63 72 50 Information 6, rue Dewoitine - BP 14 Email: plessis@matrams2i.fr F-78142 Vélizy-Villacoublay Participants MATRA Systèmes et Information University of KENT UNISYS Ltd (Phase I Only) UNISYS Payment System Ltd (Phase I Only) EMPIRE Stores Plc (Phase II Only) Country F Role C UK UK UK P P P UK P Start Date February 1996 Duration 39 months This page is located at www.cordis.lu/esprit/src/21050.htm It was last updated on 29 October 1998, and is maintained by Patrick.VanHove@cec.be Octobre 1998 High Performance Computing and Networking / 47 21073 3DULTRA High Performance Computing for Clinical Applications of 3D Ultrasound Keywords: High Performance Computing and Networking Technical - Embedded Systems, Parallel Platforms, Three-dimensional Ultrasound Uses - Computer Graphics, Real-time Applications, Visualisation Industrial Applications - Medical Diagnosis and Imaging Three-dimensional ultrasound has the potential to become a new, fast, non-radiative, non-invasive and inexpensive tomographic medical imaging technique. The medical relevance of threedimensional ultrasound has already been proven for a wide variety of routine clinical applications related to foetal examination, arterial plaque measurement, gastroenterology, abdominal tumours, cardiology, aneurysm, infant hip measurement and arthritis. On a conventional system the computing time required for a three-dimensional reconstruction is too long varying from several minutes to a few hours for each image. These times are not acceptable for routine clinical applications and hinder general clinical use of the method. The goal of the project is to develop a three-dimensional ultrasound system suitable for routine clinical applications. The necessary reduction in processing time will be achieved through the use of a high-performance computer system for the reconstruction of three-dimensional volumes. The involvement of clinical practitioners in the project ensures the market relevance of the results. Contact Point Arthur Gritzky Kretztechnik AG Research and Development Tiefenbach 15 A-4871 Zipf tel: +43 7682 2261 0 fax: +43 7682 2261 47 email: gritzky@kretz.co.at Participants Kretztechnik Fraunhofer IGD Parsytec Computer Graphi Media HCU Valencia Hopital St Louis, Paris KPGF Mainz Klinikum der Stadt Mannheim Country A D D GR E F D D Start Date January 1996 Duration 24 months Octobre 1998 Role C P P A A A A A High Performance Computing and Networking / 48 21089 HIPERTEX High Performance System for Textile Evaluation Keywords: High Performance Computing and Networking Technical - Embedded System, Data fusion, Digital Signal Processors, Distributed computing, Image Processing, Neural Network, Object Oriented Design, Parallel Architectures, Pattern Recognition, Real-time Control, Real-time systems, Time-critical Applications Uses - Inspection Systems, Process Control, Quality Control Industrial Applications - Real-time Applications, Textile Production, Vision Systems The aim of the project is the design and realisation of an embedded parallel real-time system for the on-line quality control of finished textile products. The project has strong potential for impact in the textile-manufacturing process by addressing one of the most difficult tasks in the implementation of a fully automatic plant. It will customise a general-purpose software and hardware platform for quality control in the production of textiles. The system to be defined will be sufficiently generic to be applicable to other industrial sectors. The project makes extensive use of HPCN technology in order to achieve the performance needed in this real-time environment. A fundamental aspect of the industrial prototype will be its cost, significantly lower than that of existing systems and compatible with the requirements of industrial SMEs. It is estimated that the potential world market for such medium-sized systems lies between 500 and 1000 installations. It is expected that an even bigger market exists for scaled-down smaller systems. Contact Point Eugenio Mandelli Legler Industria Tessile SpA Via San Clemente, 53 I-24036 Ponte San Pietro tel: +39 35 609035 fax: +39 35 609000 email: 101501.2666@compuserve.com Participants Legler Industria Tessile Istituto Elaborazione Informazione /C.N.R. maVis Parsytec Computer Scriba Country I I Role C P D D I P P P Start Date April 1996 Duration 24 months Octobre 1998 High Performance Computing and Networking / 49 21102 EUROPORT-D PROJECT European Porting Action - Dissemination Keywords: High Performance Computing and Networking Technical - Massively Parallel Computers, Parallel Processing, Simulation, Workstation Clusters Uses - Animation, Combustion and Explosions, Complete Flow Analysis In and Around Cars, Crash and Crash Safety, Drug Design, Electromagnetic Compatibility, Extrusion of Plastics, Forging, SAR Processing Industrial Applications - Agriculture, Cartoons, Chemical and Pharmaceutical Design, Construction, Manufacture of Automobiles and Aircraft, Oil and Gas Production, Radiotherapy, Telecommunications, Traffic Simulation The project focuses on the application of previously parallelised codes to real industrial and commercial problems. The experience gained through this and from the EUROPEAN PORTING PROJECTS is used as the basis for an ambitious and wide-ranging dissemination campaign aimed at convincing industrial and commercial endusers that HPCN is a relevant technology. This campaign emphasises the real business benefits and competitive advantage which HPCN can provide in many industrial sectors. The project builds on the results of the highly successful EUROPEAN PORTING PROJECT-1 and EUROPEAN PORTING PROJECT-2 in which some 38 industrially and commercially relevant codes were ported to parallel architectures with striking results. Ten so-called deployment sub-projects have been selected from a variety of areas including such diverse applications as cartoon animation and fire and explosion simulation. Within each of these sub-projects a parallelised code has been deployed into end-user organisations and the effectiveness of the new technology measured, using metrics focused on the end-user's business. Contact Point Owen Thomas tel: +49 2241 14 2330 GMD-SCAI Schloß Birlinghoven D-53757 Sankt Augustin Participants GMD-SCAI Smith System Engineering Start Date June 1996 Octobre 1998 fax: +49 2241 14 2102 email: europort@gmd.de http://www.gmd.de/SCAI/europort/ Country D UK Role C P Duration 17 months High Performance Computing and Networking / 50 21111 HPC STANDARDS HPC Standards Keywords: High Performance Computing and Networking Technical - High Performance Fortran, Message Passing Interface, Parkbench, Benchmarks Use - Benchmarking The aim of this project is to provide European input to and dissemination of the development of three major international standards in the area of HPCN, namely High Performance Fortran (HPF2), the standard Message Passing Interface (MPI-2), and a standard set of parallel kernels and benchmarks (Parkbench). These standardisation activities take place mainly in the USA, and in order to influence their outcome Europeans need voting rights. This requires attendance at meetings in the USA and elsewhere. In this project a fixed team, chosen for their expertise in appropriate fields, have participated in meetings of the relevant standardisation committees, namely the HPF Forum (HPFF), the MPI Forum (MPIF) and the Parkbench group. The project has ensured the early dissemination of the emerging HPF-2, MPI-2 and Parkbench standards to European industry and academia, and enabled feedback from them. Workshops have been arranged to disseminate information on relevant tools being used and developed by academia and industry. These workshops, open to all interested parties, included speakers from Europe and the USA. Contact Point Anthony Hey Department of Electronics and Computer Science University of Southampton UK - SO17 1BJ Southampton Participant University of Southampton Start Date June 1996 Octobre 1998 Country UK tel: +44 1703 592748 fax: +44 1703 592978 email: ajgh@ecs.soton.ac.uk http://www.ecs.soton.ac.uk/Projects/hpc-stds Role C Duration 18 months High Performance Computing and Networking / 51 22503 EROPPA Experiments on Remote Operations for Post Production Applications Keywords: High Performance Computing and Networking Technical - Job Management, Metacomputing, Frame Rendering and Transmission, Scene Description Language Uses - User interfaces, Three-dimensional Graphics, Post-Production Industrial Applications - Advertising, Animation, Television Post-production companies use three-dimensional graphics applications to produce film and video footage for commercial advertisements and television productions. The computing time for such applications takes several weeks on conventional computers, but could be reduced to hours through the use of high-performance parallel systems. The objective of EROPPA is to design, implement and test methodologies for remote access to computationally intensive three-dimensional graphics applications running on high-performance facilities for use in post production. Such access, if effective and economically viable, would change the post-production business dramatically. The project is driven by the real market requirements and needs of companies in this area, particularly SMEs who cannot afford to buy higher end computers. A major goal of the project is to demonstrate the use of high-performance computer systems available via high-bandwidth networks to provide an economic option for EU SMEs working in the post-production area. Such a development could have a significant impact on their competitiveness with the additional benefit of enabling more productive multi-site working Contact Point Dora Varvarigou National Technical University of Athens Heroon Polytechniou 9 str. GR-157 73 Zographou Athens Participants Country Mangos Studios GR NTUA GR Eindhoven University NL Raster Facilitair NL Genias Benelux NL Start Date October 1996 Octobre 1998 tel: +30 1 7722 533 fax: +30 1 7722 534 email: dora@telecom.ntua.gr http: to be announced Role C A A A A Duration 12 months High Performance Computing and Networking / 52 22505 WARPSPEID Wide Area Project Supporting Pan-European Information Distribution Keywords: High Performance Computing and Networking Technical - ISDN, Frame Relay, Multimedia Databases, TCP/IP, World Wide Web Uses - Component Supply, Technical Support Industrial Applications - Catalogue Marketing, Electronic and Mechanical Engineering Electronic component suppliers have large image databases holding technical support information and product data sheets. The core size and number of weekly updates make it difficult for such suppliers to maintain these databases over a number of international sites. A centralised approach with conventional telecommunications links is infeasible as access and transfer time would prevent practical use of the system for support engineers in direct telephone contact with customers. The interested audience for the results, and the associated public deliverables, will be from a number of areas. • From distribution or manufacturing industry for any market sector particularly where there is a need to provide customers with good quality, presentable material for customer support. • Large organisations wishing to provide wide read only access to corporate, product or project documentation over an Intranet, where per-seat licensing of a DMS solution would be cost prohibitive. • Organisations such as educational establishments or research institutes wishing to provide on-line access to a reference library, via Intranet, Extranet or Internet. Following the successful completion of the project, a number of key results have been identified. Offering an on-line information service in place of or alongside a telephone agent mediated service requires some change to the processes in place. The information gathering and processing methods require amending to suit the chosen technology and the potential audience for the information. The WARPSPEID project has demonstrated that Dynamic web servers are able to mediate controlled access to data from a broad range of sources ranging from legacy mainframe systems to state of the art object databases. These techniques can be deployed to provide solutions for many business processes. The Extranet capabilities used for the limited deployment to customers within the WARPSPEID project would be ideally suited for an organisation to deploy to a moderate number of business partners, such as major customers or suppliers, in order to deploy applications that enhance supply chain communication or reduce transaction costs. Other results, conclusions or impact included: • Browser based applications can be developed to access existing system capabilities • Direct involvement of users by using static HTML scenarios in conjunction with requirements definition and tools reduced development time • Security is a key issue for any application that will enable additional connections to the corporate network. • The frame relay based WAN configuration used by the WARPSPEID project proved very successful for the project. The burst and extended information rate characteristics of the frame relay service worked well with the application and was able to support multiple users on the remote sites without immediate deterioration in service. • ISDN remote access are a practical cost effective means of connecting remote locations where there is regular but non-continuous use of the service. • For large numbers of remote clients it is worth considering the use of the Internet if the service levels available are adequate, and the security risks are adequately addressed. • The WARPSPEID project has proven that it is not necessarily simple to provide a connection to a remote LAN, even when using a remote access solution from a service provider Octobre 1998 High Performance Computing and Networking / 53 The advantages of contracting out an organisation’s WAN, to a service provider, are widely acknowledged to be beneficial and far outweigh the disadvantages. However there are some disadvantages which are worth noting. Key amongst these, for a project like WARPSPEID, are: • The solutions deployed by a service provider are often only a subset of those available to the market • Newer technologies may be adopted slowly by the supplier, as the technology has to be proved against the whole of the supplier’s network before release to customers. The RS Components Internet trading channel (http://rswww.com/), which was developed during the time of the WARPSPEID project, offers an on line technical support service developed from WARPSPEID. The system architecture deployed for this was substantially different relying upon separate severs and firewalls to ensure the protection of the group wide network. A fuller presentation of results is available in PDF format from http://www.mari.co.uk/warpspeid/proj1.htm#Results. Contact Point Phil Fisher MARI Group Ltd., Computer Systems Division Old Town Hall UK – NE8 1HE GATESHEAD Participants RS Components MARI Group Ltd. Lee Dickens Ltd. Jacob Hatteland Supply Mulder-Hardenberg North Down and Ards Institute Start Date November 1996 tel: +44 191 402 0191 fax: +44 191 402 1112 email: phil.fisher@mari.co.uk http://www.mari.co.uk/warpspeid/default.htm Country UK UK UK S NL UK Role C A A A A A Duration 19 months Completed July 1998 This page is located at www.cordis.lu/esprit/src/22505.htm It was last updated on 12 October 1998, and is maintained by Merce.Griera-I-Fisa@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 54 22508 MSC Multi-Site Computing Keywords: High Performance Computing and Networking Technical - High Performance Networking, Parallel Computing, Multi-Site Application, Relational Database Uses - Process Control, Simulation, Planning, Optimisation, Multimedia Education Software, Cooperative Work Industrial Application - Environment, Software Consultancy, Multimedia Software The objectives of MSC were to experiment, to evaluate and to promote the use of HPCN for Industry and Education. A networking infrastructure was established using IP over frame-relay and intelligent networks, ISDN access and ATM. At the end of the project emerging cable-TV networks were experimented. Commercial and proven GroupWare and videoconferencing applications were used to support collaboration and remote work. Limitations in the underlying IP service provided forced the use of unicast based applications. In the project there were three main different users groups: Industry, Education and Entertainment. Industry: One of the main goals of MSC was to facilitate HPC access to SMEs through ISDN. The sub-projects were selected from companies using HPC at the European Centre for Parallelism of Barcelona (CEPBA) located at the Universitat Politècnica de Catalunya (UPC). AZTI, EyS Consulting and NEOSYSTEMS were selected at the beginning. CASA, Quantisci, Hubelsa and Cebal Entec were added during the project execution. Education: The Clear Air Project involving 14 schools in The Netherlands and Spain -and indirectly the whole I*EARN community- was selected at the beginning. Universitat Oberta de Catalunya (UOC) and a collection of experiences in the Bages country were added during the project lifetime. Entertainment: Teatre Malic is the smallest theatre in Barcelona and with the creation of the website it has become the biggest theatre in the world. "Calidoscopi" was a weekly radio program addressed and with the participation of teenagers. Macromedia Games have been successfully build and distributed over the Internet. Espais Telemàtics SL an SME specialising in the development of networking multimedia projects was the project co-ordinator. Contact Point Narcis Vives Espais Telematics, S.L Diputacio, 260, 6, 2 E - 08007 Barcelona Participants ESPAIS TELEMATICS S.L. UPC COMCOOP AZTI UOC MALIC NEOSYSTEMS Start Date September 1996 tel: +34 93 302 55 43 fax: +34 93 412 27 26 email:nvives@pangea.org http://msc.pangea.org Country Role E C E A E A E A E A E A E A End Date May 1998 This page is located at www.cordis.lu/esprit/src/22508.htm It was last updated on 1 July 1998, and is maintained by Merce.Griera-I-Fisa@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 55 22509 ESD-NET High Performance Electronic System Design Network Keywords : High-Performance Computing and Networking, Technical : Computer Supported Collaborative Work, Primary Rate ISDN, ATM, Fast Ethernet, Distributed X Windows Uses : Distributed Electronic, Mechanical Design and Simulation Industrial Applications : Telecommunications, Digital PBX, ISDN, DECT The specialised electronics business is a high technology endeavour where a unique and highly complex product must be designed and optimised while using sophisticated ESD and simulation tools, a long chain of suppliers, customers and silicon foundries and the many constraints of the global open market. The proposed project intends to implement on a small scale a distributed working environment, based on high performance networking infrastructure readily available, i.e. Primary Rate ISDN and ATM switching. The distributed environment will comprise of the main manufacturing facility, the suppliers chain and the various specialised design teams. The objective will be to demonstrate, on the basis of an existing design and production contract for a telecommunications product, the benefits associated with the use of HPN in the area of Electronic System Design. The project will use existing commercial or beta tools for distributed simulation, modelling and design, on a common ESD platform exploiting the ISDN services for European level connectivity and ATM services for current in-house and prospective broadband connectivity. Contact Point Manuela Schnuerer tel: +49 711 1398 13-0 SOLINET GmbH fax: +49 711 1398 13-999 Mittlerer Pfad, 26 email: M.Schnuerer@Solinet.com D-70499 STUTTGART http://www.solinet.com/esd-net/ Participants Solinet GmbH NCSR "DEMOKRITOS" LIC/ESL - FORTH I&E PAPADAKIS ltd MICRELEC TELETEL Country D GR GR GR GR GR Start Date September 1996 Duration 21 months Octobre 1998 Role C A A A A A High Performance Computing and Networking / 56 22517 VISTA Virtual Interactive Studio Television Applications using Networked Graphical Supercomputers Keywords: High Performance Computing and Networking Technical - Virtual Reality, Image Processing, Real-time Interactive Studio Applications Uses - Audiovisual and Computer Graphics, Interactive Television Programmes Industrial Applications - Broadcasting and Television, Education The main objective of the project is the integration and demonstration of a generic and extensible system to support the generation of scenarios for real-time interactive television. This realisation of a virtual interactive studio will offer facilities for mixing real and synthetic objects and characters in any combination, to be controlled remotely by end users such as home television viewers. The system will enable the realisation of innovative interactive domestic tele-services. The image content of the programmes will be generated over ATM in real-time by high-end graphical computers. To reduce costs telephone networks will be used for viewer feedback. The number of simultaneous interactive viewers will vary from a few to several hundred. Three concrete and specific applications will be tested using the system: •An interactive television drama (simultaneously on three channels) with viewer participation •Creation of a virtual presenter who can be controlled remotely in real-time. •Creation of an interactive virtual driving environment to teach traffic rules. For the first time broadcasting companies will be able to transmit interactive programmes in which viewers can actively direct and participate. The project will build on the expertise and technology of the HUMANOID, HUMANOID-2 , CHARM, VISINET and VLNet projects. Contact Point Joao Mena de Matos European Design Centre b.v. P.O. Box 6279 NL-5600 HG Eindhoven Participants European Design Centre BRTN EPFL Limburgs Universitair Centrum NOB Interactive University of Bradford University of Geneva VPRO Televisie Androme Start Date August 1996 Octobre 1998 Country NL BE CH BE tel: +31.40.2466350 fax: +31.40.2466850 email: menadematos@edc.nl http://www.edc.nl/vista/index.html Role C P P P NL UK CH NL B P P P P A Duration 30 months High Performance Computing and Networking / 57 22519 RCNET Exploiting HPCN in an Engineering Consultancy Environment Keywords: High Performance Computing and Networking Technical - High Performance Networking, ATM, Remote Engineering Simulation, Industrial Applications - Blast Impact, Explosive Loading, Automobile Simulation Reverse Engineering Ltd (REL), an engineering consultancy firm using numerical simulation tools, have obvious HPCN requirements, but are unable to justify the large capital costs of purchasing high end equipment. Typical uses are the numerical simulations of the responses of structures to blast impact and explosive loading. Applications for this work include the development of side impact safety systems for vehicles, and the qualification and decommissioning of offshore structures. The main objectives of the project are: • To show how existing hardware resources can be utilised more efficiently by exploiting high performance networking. • To investigate the usefulness of European broadband networks for improving working practices and increasing the throughput of work. • To investigate the usefulness of visualisation environments for embedded computational steering of remote simulations. The results of the experiments will be reduced execution times for existing numerical simulations and the ability to run larger simulations which more accurately capture the physics of the problems. Through dissemination and training activities, the results of the project will promote the use of HPCN amongst engineering firms within Europe encouraging them to improve their competitiveness. In addition, through REL’s client base the results will be disseminated into different market sectors throughout Europe. Contact Point Paul Jacob tel: +44 161 228 32 10 REVERSE ENGINEERING fax: +44 161 228 32 11 2nd floor- Brancaster Road email: jacobp@relman.demon.co.uk UK-M1 7ED MANCHESTER http://www.man.ac.uk/MVC/research/rcnet/ Participants REL-(REL Manchester) REL-(REL Norway) MANCHESTER UNIVERSITY Start Date: September 1996 Octobre 1998 Country UK N UK Duration 24 months Role C A S High Performance Computing and Networking / 58 22528 CANET Collaborative Automotive Network Keywords: High Performance Computing and Networking Technical - High Performance Networking, ATM, Multimedia Uses - Electronic System Design, Cooperative Work, Computer Aided Design, Distributed Design Industrial Applications - Car Industry The purpose of the project is to set up user experiments on the usage of computer supported collaborative working (CSCW) tools over an ATM wide area network in a distributed environment. The experiment is a user-driven trial which will involve leading edge automotive manufacturers and suppliers in a multi-site topology, using high performance communication services and applications. Some of the involved companies were early adopters of such advanced communications and development tools and are now willing to extend their expertise in this domain. The objective is to qualify and to assess the viability of such advanced tools on both technical and economical standpoints through field trials involving real users in each industrial company : Renault and Siemens Automotive in 4 different locations. These field trials are now possible since ATM technology, services and products are becoming available and advanced collaborative working tools are being developed and tested. The whole automotive business process is to be impacted by the results of the trial: impact on the decision-making process before launching a new vehicle, impact on experimenting best practices, impact on the time to market and the competitiveness of the European automotive industry, impact on the development costs of a new vehicle. Contact Point Michèle BRUN-CROZET tel: +33 5 61 19 85 16 Siemens Automotive S.A. fax: +33 5 61 19 25 17 Avenue du Mirail, BP 1149 email: mbrun@scn.de F - 31036 TOULOUSE Cedex http://www.expertel.fr/CANET/ Participants Country Role SIEMENS Automotive S.A. F C Renault S.A. F A Siemens Bereich D A Automobiltechnik MECONZA E A FT-EXPERTEL F A LAAS-CNRS F A Start Date Duration September 1996 18 months Octobre 1998 High Performance Computing and Networking / 59 22542 COVAS Collaborative Visualization of Automotive Simulations Keywords: High Performance Computing and Networking Technical - Cooperative Working, Distributed Systems, Mechanical Engineering, Networking, Simulation, Visualisation Industrial Applications - Automotive Design, Crash and Engine Simulation The project aims to evaluate the effectiveness of the use of a collaborative working environment in the design of a new car. This environment will be based around multimedia communication tools and will draw on experiences already gained in a relevant project involving ARUP and Falkenburger und Partner. The objective is to set up a multi-site high-performance computing environment and to determine its cost-benefits when applied to a real engineering problem from the automobile industry involving several collaborating partners operating under realistic commercial constraints. A further objective is the establishment of best industrial practices for such a form of working. Although drawing on an example from the automobile industry, the results of the project will be applicable to many other industrial problems involving engineering simulations and collaborative working. The COVISE system developed by RUS (subcontractor to Porsche) will form the basis for the work of the project. This system includes dedicated data management and support for several simultaneous users of high-performance visualisation. Engine and crash simulations will both be supported by this environment and will enable engineers from different organisations to work together on a common design. Contact Point Christoph Gümbel tel: +49 7044 35 4045 Dr. Ing. H.c.F. Porsche AG fax: +49 7044 35 3104 Porschestrasse email: guembel@porsche.de D-71287 Weissach http://www.netapdex.infowin.org/ Participants Porsche AG Falkenburger und Partner ARUP Start Date: November 1996 Octobre 1998 Country Role D C D A UK A Duration 24 months High Performance Computing and Networking / 60 22556 APAT Applications Professionnelles sur ATM -APAT Keywords: Cooperative Design, multi-media platform, ATM network Technical - Distant training, process control systems Industrial Applications - Nuclear plants The APAT project deals with professional applications based on new telecommunication technologies in research and industry. These multimedia technologies which incorporate video, audio and high level graphics over IP or ATM networks, based on new emergent standards (HTML, Java, VRML) allow each engineer to exploit remote distributed multimedia databases, and share know how, • accessing to remote resources, • collaborating with remote teams, • sharing data, pictures, video, audio. Possible applications are illustrated in two domains. The first one is related to the demonstration of the efficiency of a distributed multimedia and co-operative training tool in a research environment. This tool is aimed to provide world wide users of a very complex simulation software in the nuclear safety field with the ability to have multimedia training or distant expertise included in the working environment. So, it is expected that this user will have a better mastery of the code and the physical phenomena he intends to simulate. The second application is relevant to the industry of tyres and cables, a highly competitive market. The goal of this second application is to experiment concurrent engineering of tyres and cables over ATM networks (LAN's and WAN's). The co-operative design of tyres will be promoted by providing the design teams who are located in Italy and Germany with the ability to share their CAD software sessions as well as multimedia training tools (videos and texts), helpdesk and design data bases. Contact Point Andre Latrobe tel: +33 4 76 85 99 00 CISI fax: +33 4 76 85 99 20 rue de l'Arcelle, 6 email: latrobea@mercure.cisi.fr F-38600 FONTAINE http://cisi.it/public/apat/apat.htm Participants CISI S.A. CISI AID S.p.A. Pirelli S.p.A. Start Date: September 1996 Octobre 1998 Country F I I Duration 18 months Role C A A High Performance Computing and Networking / 61 22559 COLBA Customer On Line Behaviour Analysis Keywords: High-Performance Computing and Networking, Financial Applications Technical - On Line Data Analysis Uses - Finance, Telecommunications Industrial Applications - Customer Behavioural Analysis The objective is the development of a core European technological offering in the emerging sector of Customer On Line Behavioural Data Analysis. The need for high performance fraud detection applications based on this core technology has evolved, with different levels of maturity, in different markets. In the electronic money market dedicated software products have already appeared in the US. In the telecommunications market the needs have more recently been identified, and consequently there is significant market potential. In the international network services arena the commercial exploitation of the Internet is introducing new access monitoring challenges for service providers. The sectors identified above share some basic common features: in all of them there is a very large customer base accessing a digital network based service and generating very high transaction volumes. In each of these sectors fraudulent/criminal behaviour is increasingly experienced, and is expected to continue its growth at the exponential rate of the Internet expansion. A combination of HPCN technologies, rule-based systems and advanced pattern recognition techniques can provide a suitable solution to the need of monitoring these transactions and detecting fraudulent behaviours in an on-line environment. The focus is on the requirements of the credit card market, where a significant market for on-line fraud detection is mature. The validity of the approach in the telecommunication sector will be assessed through a set of focused experiments. A rule-based and neural network based fraud detection software prototype will be developed and a performance characterisation activity, on SMP architectures, will be executed with specific reference to credit card fraud detection requirements. In addition, experimental activities on telecommunications frauds and a market study in the relevant sectors will be performed. Contact Point Paolo Stofella tel +39 2 280141 Artificial Intelligence Software fax +39 2 610853 Via Esterle 9 email:paolos@ais.it I-20132 Milano Participants Artificial Intelligence Software Deutsche Bank University of Southampton-PAC Stratus Computer Ireland Energis Communication Start Date June 1996 Octobre 1998 Country I I UK IRL UK Duration 18 months Role C P P P P High Performance Computing and Networking / 62 22684 EPOCH Electromagnetic Product Optimisation using Computers with High performance Keywords: High Performance Computing and Networking Technical - Parallel Program Development, Numerical Applications, CAD, Optimisation, Tools Uses - Electromagnetic Simulation, Electrical Engineering, Electronic Engineering, Machinery, Electrical and Optical Equipment The overall objective of EPOCH is to develop a prototype software suite for designers of electromagnetic products. The prototype will allow designers to achieve optimum electromagnetic design of these products. The software will be implemented on parallel computers and clusters of PCs, taking advantage of the inherent parallelism that exists within some optimisation procedures, and previous experience in parallel implementations of computational electromagnetics software. At month 18 of the project, a fully functional 2D optimisation suite has been delivered and the last 6 months will be used to make progress towards 3D optimisation. Applications of the software within the project include optimised design of motors, video recorders, microwave ovens, television tubes and accelerator magnets, but the tools will be of a sufficiently general nature as to be exploited in many other areas. Contact Point Chris Riley Vector Fields Ltd. 24 Bankside Kidlington UK-OX5 1JE Oxford Participants Vector Fields Oxford Parallel Labein Nederlandse Philips Bedrijven Groupe Moulinex Ansaldo Energia Universita' di Genova Ecole Central de Lille Start Date January 1997 tel: +44 1865 370151 fax: +44 1865 370277 email: cpr@vector.demon.co.uk http://www.comlab.ox.ac.uk/oucl/oxpara/epoch/home.htm Country UK UK E NL Role C P P P F I I F P P A A Duration 24 months This page is located at www.cordis.lu/esprit/src/22684.htm It was last updated on 12 October 1998, and is maintained by Agnes.Bradier@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 63 22685 MESODYN Software for Mesoscale Dynamics Simulations on HPCN Platforms in Chemical Engineering Keywords: High Performance Computing and Networking Technical - Simulation, Computational Chemistry, Mesoscale chemical engineering Users - Petro- chemical industr y, Industrial Applications - Engineering of complex polymer liquids The MesoDyn project bridges the gap between micro- and macroscale modelling by the development of a new class of models and software modules for the mesoscale chemical engineering of complex industrial polymer liquids. The basic process to be modelled is the (micro) phase separation dynamics of a complex polymer liquid, including the influence of boundary conditions, hydrodynamic interactions and reactio-diffusion-phenmomena. Functionalities will include methods for geometry constraints, fluctuating convection and deterministic sheering currents, compressibility modelling, reaction-diffusion modelling, electrostatics for systems at higher ionic strengths, parameterisation tools as well a graphic user interface and appropriate analysis tools. The GUI tools will be implemented into the molecular modelling software package Cerius² of Molecular Simulations Inc. Results will be validated on industrial physical systems of the end users of MesoDyn, i.e. by BASF, Norsk Hydro and Shell Chemicals, covering the most important material classes in which mesoscale morphology has a high impact on the final product properties. Contact Point Anne Schreieck BASF-AG SZ/ZC Systems for Chemistry Carl Bosch Strasse C13 D-67056 Ludwigshafen Rheinland-Pfalz Participants BASF Shell NH MSL IBM UOG Start Date November 1996 Octobre 1998 Tel: +49 621 60 78 253 Fax: +49 621 60 49 463 E-mail: anne.schreieck@zx.basf-ag.de Country D NL N UK D NL Role C P P P A A Duration 36 months High Performance Computing and Networking / 64 22689 SIMTRAP Simulation of Traffic induced Air Pollution using Parallel Computing in a Distributed Network Keywords: High Performance Computing and Networking Technical - Parallel Computing, Simulation, Distributed Networks Uses - Dynamic Traffic Flow Modelling, Air Pollution Modelling Industrial Applications - Transportation and Environmental Planning Environmental impacts of traffic are of major concern throughout many European metropolitan areas. There is a lack of comprehensive and fast tools to estimate the level of pollution in respect of varying traffic flow conditions to optimise remedial measures. Within the project such a tool will be developed and applied at four test sites. An existing traffic flow model will be parallelised to reduce computational burdens. The system will be coupled with an existing air pollution model on a Geographical Information System environment. The integrated simulation tool is meant as an optimisation and decision support tool for transport planners and traffic engineers. The decision support system will be applied in the metropolitan areas of Berlin, Milano, Maastricht and Vienna. Various traffic control strategies and the impact on traffic flow and pollution will be tested. The test sites will function as demonstration cases for further exploitation of the developed decision support system called SIMTRAP. Contact Point Klaus Noekel PTV system Software und Consulting GmbH Stumpfstrasse 1 D-76131 Karlsruhe Participants Country PTV system GmbH D GMD-First D ESS A Hague Consulting Group NL Unseld + Partner A Provincia di Milano I Start Date December 1996 Octobre 1998 tel: +49 721 9621 200 fax: +49 721 9651 399 email: noe@system.ptv.de Role C P P P P P Duration 30 months High Performance Computing and Networking / 65 22691 SEEDS Simulation Environment for the Evaluation of Distributed traffic control Systems Keywords: High-Performance Computing and Networking, Technical - Simulation, Distributed Systems, Networks Uses - Traffic Monitoring Industrial Applications - Air Traffic Control,Maritime Traffic Control,Sensors Integration In this Project, a distributed interactive Simulation Environement (SE) for the analysis and evaluation of distributed traffic control systems will be designed, set-up and tested. The SE will provide a general purpose tool kit for the simulation of a wide range of industrial applications of strategic relevance, such as air and maritime traffic control systems. In the frame of this Project, the application area chosen as a case study is the ground traffic control in airports, that is the A-SMGCS (Advanced Surface Movement Guidance and Control System). An A-SMGCS consists of the provision of guidance, control and regulation of all aircraft and ground vehicles in the movement area of an airdrome. Such a system has to maintain the runway capacity in all weather and traffic conditions, assuring a high level of safety thus improving (with automatic means) the present visual-voice communication situation. The SE will be based on a network of commercial high performance workstations (HPCN) playing the roles of the different actors (traffic controllers, aircraft pilots, vehicle drivers, external world) and of the different sensors (radar, camera, GPS, eye) used in the A-SMGCS. Contact Point Sebastiano Bottalico tel +39-6-4150 3160 ALENIA - Area Sistemi fax +39-6-4150 3709 Via Tiburtina Km 12.400 email: sbot@lti.alenia.it I - 00131 Roma http://www.lti.alenia.it/ep/seeds.html Participants Alenia Sicta Sogitec Rigel Engineering University of Siena Sogel/Lux SEA University of Münich Country I I F B I L I D Start Date January 1997 Duration 30 months Octobre 1998 Role C P P P A A A A High Performance Computing and Networking / 66 22693 HYPERBANK High Performance Banking Keywords: High Performance Computing and Networking Technical - Business Knowledge Modelling, Data Warehousing, Data Mining, Parallel Computing Uses – Generic Models for Customer Profiling Industrial Applications - Banking, Finance, Insurance, Retail, Utilities The main aim of the project is to provide the Banking sector with the requisite toolset for increased understanding of existing and prospective customers, and better tailoring of products and services for those customers. The entire ethos of the project will be user led. The successful completion of the project will allow banks to gain competitive advantage and specifically enable them: • to offer the right set of products for each customer, at the right price for each customer • to ensure maximum retention of customers • to ensure maximum efficiency in obtaining new customers • to manage the risk to the bank from each contract. The approach integrates and deploys three important technologies: business knowledge modelling, data warehousing and data mining, and high performance computing. In exploiting these technologies, the project will deliver the following major results: (1) a generic business knowledge model of "customer profiling" within the banking sector, that can be used for different banking methods and strategies; (2) an enhanced data extraction tool incorporating the business model; (3) a high performance data mining tool interfacing to both the data extraction and the business knowledge model. Contact Point John McGuire Datel Advanced System Limited Datel House Caspian Road, Atlantic Street Altrincham, Cheshire WA14 5HH Participants Country Datel UK NBG GR Postgirot S Capital Bank UK 01-Pliroforiki GR Carleton Europe B UMIST UK KTH S Start Date January 1997 tel: +44 161 929 6444 fax: +44 161 929 5444 email: jpm@alt.datel-group.co.uk http://www.co.umist.ac.uk/CompISG/ISE-Projects/Hyperbank.html Role C P P P P P A A Duration 30 months This page is located at www.cordis.lu/esprit/src/22693.htm It was last updated on 10 October 1998, and is maintained by Merce.Griera-I-Fisa@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 67 22695 SLOEGAT Short and Long Term Optimization of Electricity Generation and Trading in a competitive Energy Market Keywords: High-Performance Computing and Networking, Technical - Simulation, Automatic Optimisation, Parallel Computing Uses - Energy Management and Monitoring Industrial Applications - Energy Trading, Electric Power Planning The project aims to develop, implement and test, on a high performance computing platform, a software system to simulate and optimise the energy generation and trading coordination planning process in large electricity generating systems, both in the short (1 day-1 week) and medium to long term (one-two years). Special consideration will be given to this process to the growing importance of the energy trading problem in a deregulated market. The target is to develop and implement suitable mathematical algorithmic approaches for a parallel high performance environment, so that the solution times for a large size case of the above problem lie in the range of: - 10-15 minutes for the short term planning problem, and - 0,5-1 hour for the medium-long term planning problem The currently available systems cannot attempt to solve the problem optimally. They only provide implementable solutions by addressing the problem as a sequence of subproblems which, on one hand require a computing effort that should be affordable with present systems and, on the other hand, do not guarantee by any means the global optimum of the solution. The proposed project addresses the improvement of both the quality (better planning, lower costs) and the efficiency (less response time) of the electric power planning and management related activities by the electric utilities. Contact Point José Luis de la Fuente tel +34-1-5776500 IBERDROLA fax +34-1-4357910 Serrano 21-4° email: jl.delafuente@ iberdrola.es E - 28080 Madrid http: to be announced Participants IBERDROLA UPC SIEMENS VEW Energie RWTH Start Date December 1996 Octobre 1998 Country E E A D D Duration 30 months Role C P P P P High Performance Computing and Networking / 68 22696 HAT High Performance Arbitrage Detection & Trading Keywords: High-Performance Computing and Networking,Finanacial Applications Technical - Real-Time Detection of Anomalies Uses - Finance Industrial Applications - Arbitrage Detection and Trading This project will develop a realistic demonstration of the capabilities of applying HPC to detecting and trading arbitrage in high-frequency financial markets which will be of vital interest to market participants. The roles of the partners in the project are as follows: London Business School in collaboration with Reuters have already developed a set of pricing relationships that are often violated in the Foreign Exchange, Equity, and Fixed income market. A comprehensive set of zero-risk arbitrage conditions will be delivered for each market to the other partners to achieve the basic objectives of the HAT project. Dresdner Bank, BNP, and Hughes Financial Analytics will take delivery of these zero-risk routines and, with each partner concentrating on one specific market, will gather market data. When parallelised routines become available, they will evaluate the areas where the new technology could provide them with added value. The IT partners, Intrasoft and Parsytec will provide the hardware and software for i) connecting to Reuters datafeeds ii) checking for integrity iii) implementing in real time iv) logging violations and v) providing interfaces with alarm generators for risk management. The institutions will paper-trade the system for a period of one year and will assess the risks involved in using it. Reuters will lead the work associated with defining user requirements. It will also take prime responsibility for the subsequent marketing of the HAT Project’s results. The deliverables will be productised and exploited using Reuter’s established marketing and distribution channels by offering a high performance product as an add-on service to Reuters’ Markets 2000 and 3000 product-lines. Contact Point Harald Wegner tel +49 69 263 16 290 Dresdner Bank AG fax +49 69 263 16 540 Jurgen-Ponto-Platz 1 email: Harald.Wegner@Dresdner-Bank.de D-60301 Frankfurt am Main Participants Dresdner Bank BNP Hughes Financial Analytics London Business School Parsytec Intrasoft Start Date January 1997 Octobre 1998 Country D FR UK UK D GR Duration 24 months Role C P P P P P High Performance Computing and Networking / 69 22700 CRITIKAL Client-server Rule Induction Technology for Industrial Knowledge Acquisition from Large Databases Keywords: High-Performance Computing and Networking, Decision Support Technical - Data Mining, Data Warehousing Uses - Customer Profiling & Service Industrial Applications - Finance, Pharmaceuticals The basic premise for the project is the real business need for tools to enable in situ data mining against large databases in a client-server environment. The technological basis for the work of the project is Attar Software’s mature, widely proven XpertRule induction technology and its newly released XpertRule Profiler client-server induction technology which is aimed specifically at the data warehousing marketplace. The objectives of the project are to: • demonstrate the potential for adding value to the info assets of organisations in different sectors through effective client-server induction from large corporate data warehouses; • develop and demonstrate an advanced client-server induction system capable of supporting efficient, effective data mining of large databases in business environments; • generate prototypes of a series of enhancements to Attar Software’s XpertRule Profiler product which are enabled by HPC technology and which are commercially exploitable during the lifetime of the project; • generate a body of large-scale data mining experiences, from both the financial and pharmaceutical wholesale sectors, which will form the basis for generic dissemination and specific marketing activities. Product technology based on the results of the project will be rapidly exploited by Attar Software both during and after the timeframe of the project. Products based on the software developed in the project will have the potential to make a significant impact on a huge worldwide marketplace which is eager to invest in demonstrable capabilities. The end-users will provide advice and guidance to Attar Software to assist the exploitation process. Contact Point Akeel Al-Attar tel +44 1942 608844 Attar Software Ltd fax +44 1942 601991 Newlands Road email: aalattar@attar.co.uk UK-WN7 4HN Leigh Participants Attar Software GEHE Lloyds TSB Group PAC University of Stuttgart Start Date January 1997 Octobre 1998 Country UK D UK UK D Duration 24 months Role C P P P P High Performance Computing and Networking / 70 22707 HP-RETAIL High Performance Parallel Computing for Medium-Sized Retailers Keywords: High-Performance Computing and Networking, Technical - Decision Support Tools, Parallel Data, Symmetric Multiprocessor Uses - Retailing Industrial Applications - Data Warehouse, Decision Support Food and non-food retailing are extremely competitive markets. Large retail groups concentrate their efforts on the exploitation of new consumer groups. They have developed information systems which deliver the necessary marketing information and also shorten the time delay between consumer demand and stock replenishment. High performance parallel computing is ultimately required for data warehouses. Present installations at the large retailing chains are based on massively parallel IT systems from US companies. Medium sized retailing companies are concerned that - although high performance computing is considered extremely beneficial for efficient consumer response tracking - these large systems are prohibitively expensive and not scalable to fit the needs of a medium sized company. By clustering several symmetric multiprocessor nodes, a high performance data warehouse computer can be built at a reasonable cost. However, there is a lack of availability of decision support tools for consumer response tracking. It is not possible to develop these tools with the resources of a single medium sized retailer. The project will (i) satisfy the decision support needs of many medium sized retailers through a scalable solution, (ii) add user value by tailoring the functionality of the tools for retailers, and (iii) build a high performance parallel data loader for collecting sales info from many stores. The result will be a scalable architecture and a set of pre-configured decision support tools that will help to improve the average trade margins not only overall but also for individual sales outlets, allow benchmarking between different outlets and help to optimize stock replenishment. Contact Point Gilles Bogo tel +33 1 30 80 37 96 Bull SA fax +33 1 30 80 79 50 Rue Jean Jaures BP68 email: G.Bogo@frcl.bull.fr F-78340 Les Clayes Sous Bois Participants Bull Allkauf FhG-IAO Information Management MEINL UNIGRO Start Date December 1996 Octobre 1998 Country F D D D A NL Duration 24 months Role C P P P P P High Performance Computing and Networking / 71 22709 FAST-SIM MPP System for simulated NMR data analysis Keywords: High-Performance Computing and Networking, Technical -Nuclear Magnetic Resonance, Magnetic resonance imaging,Quadrics computer Uses - Visualisation, Simulation, Computer graphics Industrial Applications - Medical diagnosis and imaging, Pharmaceutical research, Biotechnology, Molecular engineering, Biochemical research NMR spectroscopy is applied to High Resolution Spectroscopy (HRS), for biochemical characterization, and to Magnetic Resonance Imaging (MRI), for medical diagnosis. HRS is strategic for the modern chemical and pharmaceutical industry. Multidimensional (MD) NMR, one of the leading techniques, has strong computational demands, forcing users to off-line processing. This hinders complete exploitation of costly experimental data, ever increasing with new-generation spectrometers. Furthermore, conventional systems do not allow multidimensional Maximum Enthropy Reconstruction (MER) in an economical time scale. MER would cut MD acquisition times by a factor of two or more, thus enabling the study of compounds so far considered insufficiently stable to be amenable to NMR characterization. MRI is an important non invasive diagnostic tool. Huge matrices are processd, for both image reconstruction and image analysis, and new techniques will push computational resources even further. Also, MRI users need intensive processing for analysis and interpretation. Computational time is a real bottleneck in this process, with significant impact on the efficiency of the health system when it comes to diagnostic users. The project will realise a parallel software package to analyse and interpret NMR data with highly advanced techniques and capabilities running on a powerful MPP platform from QSW. The main goal of the development is to adopt software packages as an analysis and verification tool, during the design phase of new NMR data analysis equipment for spectroscopy and imaging. This allows to analyse simulated data derived from the design of novel high-performance systems for the study of biological and synthetic samples, in the chemical and the biomedical fields. A version of the package will then be derived for commercial exploitation as a stand-alone software tool, to be adopted by users of installed NMR systems. Contact Point Dr Giovanni Bizzaro Bruker Spectrospin Italiana Via G. Pascoli 70/3 I-20133 Milano Participants BRUKER QSW SIGEDA Università di Verona tel: +39 2 70 63 63 70 fax: +39 2 23 61 294 email: gb@bruker.it http: to be announced Country Role D C UK P I P I P Technische Universität Munich Start Date January 1997 D Duration 26 months Octobre 1998 P High Performance Computing and Networking / 72 22716 DoRo Document Routing Keywords: High Performance Computing and Networking Technical - Parallel Platforms, Parallel Parsing, Integration Users – Computer Document handling, Automatic indexing Industrial Applications – Document Management, Document Analysis and Workflow A problem faced by a growing number of companies in the Information Age is the routing of human-readable documents that are in electronic form. Such documents can either be received in the form of electronic messages, or they can be the result of OCR/ICR-processing of paper mail and documents. How can it be assured that they arrive at the proper destination with the smallest possible time delay? For properly addressed mail this is just a matter of logistics, but for generically addressed mail and for documents without specific addressing information, this involves some form of human intervention: reading the document, interpreting its contents, and forwarding it to the person or department that should deal with it. Manual mail routing is error prone and time consuming. How can it be automated? The purpose of the DOcument ROuting project is to develop a platform-independent system performing automatic routing of human-readable documents in electronic form on the basis of an analysis of the contents of the documents and knowledge of the characteristics of the possible destinees. The system will be embedded into standard workflow environments and will appear as a “black box component“ of such systems. It shall be generic, both with respect to the application domains and with respect to the natural languages used. Versions of the linguistic classification system will be developed for three Community languages (Dutch, Spanish, and Greek) and for three users (an insurance company, Dutch, an IT-sector company, Spanish, and a stock exchange, Greece). Contact Point Casper Molmans Xerox Professional Services Gondel 1 1186 MJ Amstelveen The Netherlands Participants Xerox PROFit Parsytec KUN CTI USC OHRA ASE IECISA Docutec Start Date January 1997 e-mail : molmansc@xps.xerox.nl http://www.profit.es/proyectos/doro-ingles.html Country NL ES DE NL GR ES NL GR ES DE Role C P P P A A A A A A Duration 30 months This page is located at www.cordis.lu/esprit/src/22716.htm It was last updated on 29 October 1998, and is maintained by Khalil.Rouhana@cec.be Octobre 1998 High Performance Computing and Networking / 73 22717 OPTIMUM Applying HPC Techniques to the Industrial Design: Automatic Optimization of Forging Processes Keywords: High-Performance Computing and Networking, Simulation Technical : Optimisation, CAOD, PVM, CAE, CAD Uses : Forge 2, Forge Industry, Training The objective of the project is to design an optimisation environment for hot metal forging based on the 2D and axi-symmetric simulation code Forge2. Parallelisation is mandatory to make such a tool exploitable in an industrial environment due to the cost of each individual simulation and the need of several intermediate steps to reach the optimum. The parallelisation can be performed in two ways. Firstly, by distributing the computation of independent trials in the case of an optimisation procedure based on the so-called iterative approximation method. Secondly, by parallelising through a domain decomposition approach, each simulation, in the case of a direct optimisation method. Most of the potential customers are SME, so the objective is to design efficient parallel tools for inexpensive systems based on networked PCs. CAOD (Computer Aided Optimum Design) is a set of methodologies of design and tools that allow automatically to obtain optimal designs starting from the product specification. The RTD work to be done during the project is: • the development of a Parallel HPC design environment for automatic optimization of forging processes (integration of CAD and CAE and interactive design), • the research and application of parallel HPC algorithms, • the training with Parallel HPC Automatic Design Tool and Dissemination of the results. Contact Point Mr. Jesus LAMA tel +34.4.489.24.00 LABEIN fax +34.4.489.24.60 Mecanica y CAD/CAM email: lama@labein.es Cuesta de Olabega 16 http: E- 48013 Bilbao FRANCE Participants FOVISA ARMINES IATRIKI TECHNIKI LABEIN FORD WERKE GSB TRANSVALOR Start Date July 1997 Country E F GR E D E F Duration 24 months Role C P P P A A A This page is located at www.cordis.lu/esprit/src/22717.htm It was last updated on 29 October 1998, and is maintained by Agnes.Bradier@cec.be Octobre 1998 High Performance Computing and Networking / 74 22723 HITERM High-Performance Computing and Networking for Technological and Environmental Risk Management Keywords: High-Performance Computing and Networking, Technical - Distributed Parallel Simulation, Visualization, Discrete Multi-criteria Optimization, Rule-based Expert Systems, Real-time Data Acquisition Uses - Time Critical Environmental Applications, Technological Risk and Emergency Management, Adaptive Routing, Simulation of Accidents, Interactive Decision Support Industrial Applications - Chemical Industries, Hazardous Installations, Transportation of Hazardous Substances, Emergency Response Management and Training Within the framework of HPCN Information Management and Decision Support, HITERM aims at expanding the application of HPCN to decision support in new domains: the central focus is the interface between technological risk management and the environment. Using distributed parallel computing, the project aims at reaching better-than real time performance for the simulation of accidental release of hazardous substances into the atmosphere and ground and surface water. This information will be used, in the framework of interactive, on-line decision support and advisory systems for the adaptive routing of hazardous transports integrating environmental risk criteria with other road information and the support of emergency management tasks (and related staff training) for transportation accidents involving hazardous substances and for hazardous installations, as foreseen by the amended Post-Seveso Directive (82/501/EEC, 87/216/EEC, COM(94)4). In addition to connecting the HPC simulations to various on-line monitoring data sources, the project will explore the on-line integration of uncertainty and error analysis based on MonteCarlo methods and on-line interactive data interpretation and visualization for dynamic, spatially distributed, and probabilistic model results. Contact Point Dr. Kurt Fedra tel +43-22 52-63 30 50 Environmental Software and Service GmbH fax +43-22 52-63 30 59 Kalkgewerk 1 email: info@ess.co.at A-2352 Gumpoldskirchen, AUSTRIA http: www.ess.co.at/HITERM/ Participants Environmental Software and Services GmbH GMD SYRECO s.r.l. LNEC Informatica SA FCCN PETROGAL Petroleos de Portugal, SA ASIT Security and Information Systems in Transport Ltd. Start Date January 1997 Country A Role C D I P P P P P P P P CH P Octobre 1998 Duration 30 months High Performance Computing and Networking / 75 22727 WEPTEL Innovative Weather Presentations on TELevision Keywords: High Performance Computing and Networking Technical - Virtual Environment, Weather Forecasting, Air Pollution, Parallel Architectures Uses - Television Presentation, Visualization, Pre and Post-processing Industrial Applications - Broadcasting The objective of the project is to introduce innovative weather presentations on television. For this purpose a visualisation system will be developed which is able to select fast and on-line interesting phenomena produced by weather forecast and air-pollution models, and combine the output of these two models. Furthermore, the system - and thereby the numerical models used in the project should be able to provide forecasts for areas varying from a regional scale to a continental scale. Finally, state-of-the-art visualisation techniques will be included in the system to introduce novel methods to present forecast phenomena ( e.g., three-dimensional data, time-series, virtual scenery). The project addresses the embedding of High Performance Computing and the post-processing of data generated by simulation into a commercial end-user application. The post-processing needs for high performance computing and related visualisation of high volume data are considered one of the major challenges for high performance computing to gain a market position beyond scientific computing. This project distinguishes itself by the fact that the post-processing needs do not cater for scientific computing and high performance computing experts but rather for an audience of nonexperts Contact Point Lex Wolters Dept. Of Computer Science Leiden University Niels Bohrweg 1 NL-2333 CA Leiden Participants NOS-News NOB TV2/Danmark KNMI DMU Leiden University Start Date November 1996 Octobre 1998 tel: +31 71 527 7054 fax: +31 71 527 6985 email: llexx@cs.leidenuniv.nl Country NL NL DK NL DK NL Role C P P P P P Duration 30 months High Performance Computing and Networking / 76 22740 PARMAT Efficient Handling of Large Matrices on High Parallel Computer Systems in the PERMAS Code Keywords : High Performance Computing and Networking Technical : Coupled Problems, Numerical Applications, Scalability, Scheduling, Software Migration Users : Structural Mechanics Simulation Industrial Applications : Energy Production and Distribution, Mechanical Engineering, Sea Transportation Vehicles Due to the experience from the EUROPORT project, PERMAS can be expected to become one of the leading Finite Element packages with respect to the utilisation of massive parallel systems. The following deliverables will come from the project: A parallel PERMAS version with good performance and scalibility for a variety of applications including multi-physics phenomena, especially of fluid/structure acoustics. An improved scheduling and clustering mechanism allowing to scale up to 64 processors or more of tightly coupled machines and up to 16 workstations connected through high speed networks. Selected case studies within offshore engineering and ship building, validated by industrial partners that demonstrate the usefulness and benefits of HPCN in terms of reduction of project time, man power and safety aspects. Publications and reports as well as seminar material to promote the usefulness and industrial relevance of HPCN for industry. Contact Point Prof. Uwe Schulz INTES Industriestrasse 2 D-70569 Stuttgart Participants INTES UPC IRCN Bureau Veritas Structural Engineering DYNALIS Start Date January 1997 tel: +49 711 7849915 fax: +49 711 7849910 email: schulz@intes.de http://www.intes.de/parmat.html Country D E F F N F Role C P P P P A Duration 24 months This page is located at www.cordis.lu/esprit/src/22740.htm It was last updated on 13 October 1998, and is maintained by Agnes.bradier@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 77 22745 INDEX Intelligent Data Extraction Keywords: High-Performance Computing and Networking, Data Reduction, Technical - Data Visualisation, Data Compression, Data Analysis, Modular Software Uses Engineering Simulation, Computational Analysis+ Industrial Applications - Electromagnetic, CFD, Seismic Applications The objective of INDEX is to make the interactive analysis of the results of engineering simulations performed on HPCN platforms practicable . The very success of HPCN methods in reducing engineering design times and costs has led to a substantial increase in dataset size and number, and the difficulty and cost of analysing and interpreting this data which (i) selectively reduces the proportion of the data returned for analysis by an intelligent and automatic preselection, (ii) automatically and transparently exercise data compression methods, and (iii) present compact alternative representations of features in vector and scalar fields. The end product of INDEX is the INDEX toolset, a portable modular software library, which will provide an integrated software solution to the problem of increased dataset size from engineering simulation. The toolset will be used by the industrial partners British Aerospace, Daimler-Benz and Osservatorio Geofisico Sperimentale within their existing design processes, and exploited externally by the partners through commercial sales of the toolset. Contact Point John Whittle tel: +44 117 936 60 93 British Aerospace (Ops) Ltd fax: +44 117 936 37 33 Sowerby Research Centre email: whittle@src.bae.co.uk FPC 267, PO Box 5 http://www.man.ac.uk/MVC/IND EX/Public/index.html UK- Filton, Bristol Participants British Aerospace Daimler Benz AVS/UNIRAS University of Manchester OGS University of Stuttgart Start Date January 1997 Octobre 1998 Country UK D DK UK I D Duration 24 months Role C P P P P A High Performance Computing and Networking / 78 22746 MEDUSA Multi-disciplinary Engineering Design via Unitary Software Applications Keywords: High-Performance Computing and Networking, Technical - Simulation, Optimisation, Parallel Systems Uses - Engineering, Design, Analysis, Multi-physics Industrial Applications - Electromagnetic simulation, Aerospace simulation The project is concerned with the multi-disciplinary design and coupled multi-physics optimisation of complex engineering products enabled by exploiting parallel High Performance Computing (HPC). The project seeks to enhance the industrial relevance of the multi-disciplinary design aspects of the EP 8328 CAESAR demonstrator project by extending it to include the detailed structural response and optimisation of complex engineering structures which are subject to coupled and combined multiphysics loads. Its two principal objectives are to establish by demonstration that HPC can enable: - the analysis of complex structures subject to multi-physics loads through realistic and detailed finite element models, so reducing the manufacturing costs committed in the design phase by at least 10 %, and - efficient coupled multi-physics optimisations, to improve product quality and the performance-tocost ratio of complex engineering products. Cost-benefits targets vary between 3 % to in excess of 10 % for the industrial end-users. These objectives will be tested against industrial demonstrators posed by the end-users in the project. Contact Point David Rowse tel +44-1179-366477 British Aerospace, Sowerby Research Centre fax +44-1179-363733 FPC 267 - P.O. Box 5 email: rowse@src.bae.co.uk UK-BS12 7QW Filton, Bristol http://www.telecall.co.uk/~srcbae/medus a Participants British Aerospace ABB Offshore Technology AVL-List GENIAS Software Computational Dynamics Resarch MacNeal-Schwendler University of Wales, Swansea Start Date January 1997 Octobre 1998 Country UK N A D UK D UK Duration 24 months Role C P P A A A A High Performance Computing and Networking / 79 22752 ACTIVATE Advanced Computing Technology for Innovative VibroAcoustic Tools in Engineering Keywords: High Performance Computing and Networking Technical - Boundary and Finite Element Methods, Multi-processor architectures, Message passing Users - Aircraft and Aerospace Design, Automotive Design, Civil Engineering Industrial Applications - Vibro-acoustic optimisation The aim of the ACTIVATE project is to develop an advanced and innovative High-Performance Computing Vibro-Acoustic Platform (HPC-VAP), which will integrate state-of-the-art HPCN ( High Performance Computing and Networking) concepts into existing well know vibro-acoustic codes. The new HPC-VAP will be based on the coupling between the structural code MSC/NASTRAN and the vibro-acoustic code RAYON-3D developed by STRACO/France, which will be parallelized in the frame of this project. RAYON-3D code is fully integrated in the CAD-CAE environment of SDRC/I-DEAS™ Master Series and incorporates advanced coupled numerical techniques based on Boundary and Finite Element Methods (BEM-FEM) for modelling vibro-acoustic radiation and transmission phenomena, including fluid-structure internal-external coupling. The ACTIVATE project will reinforce the leading position of European SME developers and vendors in the strategic market niche of vibro-acoustics, and will increase the competitiveness of European transportation industry by providing them with an advanced HPC Vibro-Acoustic Platform allowing them to drastically reduce the vibro-acoustic optimization costs, during the vehicle design phase. Additionally the vehicle comfort and the environmental quality will be enhanced. Contact Point Franck Choukroun STRACO SA 20, Rue du Fonds Pernant Z.A.C de Mercières - Zone III F-60471 Compiègne Cedex Participants STRACO CASA MSC ONERA PSA RUS AEROSPATIALE (AST) DORNIER UTC Start Date November 1996 Octobre 1998 Tel: +33 3 44 30 43 60 Fax: +33 3 44 86 87 77 E-mail: choukrou@straco.fr Country F E D F F D F D F Role C P P P P A A A A Duration 24 months High Performance Computing and Networking / 80 22765 DESIRE II Design by Simulation and Rendering on Parallel Architectures II Keywords: High Performance Computing and Networking Technical Uses Industrial Applications The project is an extension of the ESPRIT project 6173 DESIRE. The DESIRE II project aims to complete the development of the DESIRE rendering software and its interface, for distributed and remote networked communication, the implementation of parallel geometric algorithms for interactive styling, and at the first implementation of a fully functional prototype system for interactive design and engineering of car bodies based on visual feedback through photorealistic rendering and through visualisation of simulated external air flow. This will provide a solid foundation for large scale industrial realisations of the digital mock-up concept. The DESIRE rendering software will be suitable for a wide range of application areas. In addition to the application originally chosen to drive the development in the DESIRE project, i.e., interactive design and engineering of car bodies, these areas will include interactive architectural design and lighting simulation, scientific visualisation, commercial digital film production for entertainment and education, digital media games production and high image quality virtual reality applications. Contact Point Rolf Herken Mental Images Fasanenstrasse 81 D-10623 Berlin Participants Mental Images BUF Crystal Magic Dassault Systems HTC WSTL Start Date January 1997 Octobre 1998 tel: +49 30 31 59970 fax: +49 30 31 599733 email: office@mental.com Country D F UK F D UK Role C P P P P P Duration 30 months High Performance Computing and Networking / 81 22835 NETAPDEX Networking Application and Dissemination Expertise Keywords: High Performance Networking Technical - Networking, Cooperative Working, Distributed Systems, Meta Computing, Distributed Graphics and Animation, Simulation, Visualisation Industrial Applications - All industrial sectors and fields NETAPDEX is a support, co-ordination and dissemination project, that promotes network usage for European industry and research organizations. It represents multiple member projects from the ESPRIT HPCN "Preparatory, Support and Transfer Activities" specifically oriented towards "High Performance Networking Best Practice and Experiments". Companies both large and small, especially those involved in advanced industrial research, are now testing a variety of network technologies with a view to enhancing their competitiveness. NETAPDEX member projects experiment with new ways of working enabled by technologies such as ISDN, ATM, etc. and the Internet on top. A NETAPDEX prime task is to transfer practical experience to the industrial sector, as well as to operators, service providers and decision-makers, in order to consolidate experience and stimulate new activities in the field of European and world wide networking. It is becoming increasingly important to prepare professional users to take advantage of the new opportunities that will emerge as a result of the full liberalisation of the telecom market in 1998. The gathering and dissemination of concrete experience gained in many projects will help other business sectors to determine their own networking needs and expected benefits. They should also smooth the transition from pilot schemes to full blown operational implementations. To establish a broad basis of experience NETAPDEX has been built to gather further input, e.g. form other networking based projects in Europe. Therefore NETAPDEX has been opening relationships with activities such as the Technology Transfer Network (TTN) and is actively seeking further contacts. Based on the gained experiences NetApdex will express organisational, regulatory and financial requirements of the industrial user community towards providers and regulators. Contact Point Ulrich Lang RUS, University of Stuttgart Allmandring 30a D-70550 Stuttgart, Germany Participants RUS Genias Mari UPC Reverse Engineering Ltd European Design Centre Siemens CISI Start Date August 1996 Octobre 1998 tel:+49 711 685 5995 fax: +49 711 678 7626 email: lang@rus.uni-stuttgart.de HYPERLINK http://www.netapdex.infowin.org/ Country D NL UK ES UK NL FR FR Role C A A A A A A A Duration 30 months High Performance Computing and Networking / 82 23174 SISCI Standard Software Infrastructures for SCI-based Parallel Systems Keywords: High-Performance Computing and Networking Technical - SCI, Software environments (MPI, PVM, POSIX Threads) Uses - Demanding parallel application software Industrial Applications - Flow simulation, image processing, data acquisition The IEEE/ANSI/ISO standard 1596-1992, Scalable Coherent Interface (SCI), defines a high-speed, scalable, bus-like interconnect to hook up nodes: processors, memories, peripheral devices, commodity Work Stations (WS) or PCs; in a high-performance parallel system. An SCI interconnect enables a standard WS or PC cluster to be operated and conveniently programmed as a shared-memory parallel machine. Given proper application programming environments, SCI provides unprecedented flexibility and performance for parallel processing on "commercial-off-the-shelf" WS or PC clusters. The SISCI project will provide programming environments for SCI-based parallel systems. The project will develop and test on industrial applications the following formal or de facto standard software environments: • the Message-Passing Interface communication library (MPI); • the Parallel Virtual Machine programming library (PVM); • a POSIX-compliant thread package that will operate across machine boundaries Contact Point Mr. Kåre Løchsen tel: + 47 22 62 70 00 Dolphin Interconnect Solutions AS fax: + 47 22 62 71 80 P.O. Box 70 Bogerud email: kl@dolphinics.no N-0621 OSLO http: http://www.dolphinics.no Participants Dolphin AEA CERN Scali University of Bergen, Parallab Rutherford Appleton Laboratory Technische Universität München EMU Unterwasserpumpen Country N D F N N UK D D Start Date April 1997 Duration 24 months Octobre 1998 Role C P P P P P P P High Performance Computing and Networking / 83 23475 RADAR High Performance Rainfall Radar Image Processing for Sewage Systems Control Keywords: High-Performance Computing and Networking, Image Processing Technical - Neural Networks, Forecasting System Uses - Rainfall Forecasting, Load Prediction for Sewage Treatment Industrial Applications - Sewage Treatment Utilities The control of sewage and drainage systems of large cities is one of the highly critical problems that has to be solved in order to reduce pollution impact on the environment. The most difficult problem that limited the automatic control in this domain is the low level of descriptive information available on the system: the geographical distribution of rainfall can not be handled by the classic approaches based on rain gauges, and only the main pipes and components of a sewage system are known by operators. Therefore deterministic modelling is impossible in practise for the real-time control. The recent emergence of new technologies using radar weather images and artificial neural networks modelling have demonstrated that significant improvement can be achieved. RADAR will implement a set of methods for forecasting flows based on radar images and on-line rain gauges, covering a large catchment area using a flexible high performance computing environment based on CapCASE++. The main benefit for the users - to be demonstrated at three sewage treatment plants - will be a better management of critical rainfall situations by choosing the optimal management strategy, i.e. reducing CSO (Combined Sewer Overflows) events and increasing the stability of biological sewage treatment processes. Contact Point Mr. Albert Mpe tel +33 1 30 15 40 00 CITI, Lyonnaise des Eaux fax +33 1 39 76 02 93 59 av. Emile Thiebaut email: ampe@citi.lyonnaise-des-eaux.fr F-78110 LE VESINET Participants Lyonnaise des Eaux APIC SA Water Quality Institute AMGA OFAI Danish Meteorological Institute VA-Verket Malmö Start Date To be announced Octobre 1998 Country F F DK I AT DK S Duration 30 months Role C P P P P A A High Performance Computing and Networking / 84 23480 FLEXSECURE Modular, Scaleable Detection System for Explosives, Drugs and Weapons in Hand Baggage and Medium Sized Parcels Keywords: High Performance Computing and Networking Technical - Embedded Systems, Parallel Platforms, X-ray Image Evaluation Uses - Real-time Applications, Image Processing, Visualisation Industrial Applications - Security X-ray Inspection The improvement of internal and travel security by automatic or semi-automatic X-ray units is a world-wide increasing demand. Systems currently in the market are either non automatic units or they are limited to high-end applications with high throughput and are ranging therefore in the upper price areas. For hand baggage control systems placed in all airports, in harbours, custom control stations, mail rooms in security critical enterprises and other security relevant locations new modular and scaleable solutions must be developed. Scaleable for coping with increasing load, modular in order to reflect specific customer needs (explosives, drugs, weapons) today and in future. The result of this development project will be scaleable X-ray inspection system for different user groups integrating standard embedded high performance computing hardware and application specific software modules. Furthermore, the technology will help to reduce the tremendous costs concerned with the manual detection of dangerous goods and will increase the detection reliability. In the end, the security on travel or in the society could be enhanced by introducing high performance computing into end-user products. The involvement of end users in the project team, namely one of the biggest European airports and an important national aviation association ensures the market relevance of the results. Contact Point Dr. Hermann Ries Heimann Systems GmbH Research and Development Carl-von-Linde-Str. 14 D-65197 Wiesbaden Participants Heimann Systems GmbH Parsytec GmbH Kjaergaard Industri Automatic Copenhagen Airports A/S AENA (Aeropuertos Espanoles y Navegacion Aerea Start Date January 1997 Tel: +49 611 9412 575 Fax: +49 611 9412 550 email: hries@heimannsystems.com Country D D DK DK E Role C P P A A Duration 24 months This page is located at www.cordis.lu/esprit/src/23480.htm It was last updated on 29 October 1998, and is maintained by Khalil.Rouhana@cec.be Octobre 1998 High Performance Computing and Networking / 85 23483 EPSIS Real-Time Analysis of Video Image Structure Keywords: High-Performance Computing and Networking, Embedded Systems, Video Image Processing, Technical -Virtual Reality, Parallel Architectures, Target Recognition, Motion Estimation Uses - Machine Vision, Three-Dimensional Graphics Industrial Applications -Television Broadcasting , Sports Sponsoring, Video Production Image processing techniques are bringing a whole range of innovative applications to the world of digital video. An early commercial application released by one of the project partners in 1995 is the electronic billboard. This image processing system is able to detect and track a real billboard or another target in the live broadcast flow of a sports event, and to insert a synthetic "electronic billboard" to replace it in the video signal. This can be used to optimise the impact of sponsoring to specific audiences. The goal of the EPSIS project is to enhance this system by launching a new research activity focused on obstacle processing. The anticipated project results will enable the insertion of a new billboard even when a player is partially masking the targeted "real" billboard, with a minimum number of constraints. Similar techniques will also be exploited in a different application, known as the virtual set. This application allows images of actors and props recorded in an appropriate studio to be superimposed on a synthetic background. Most virtual set solutions available today require instrumented cameras to track their location and orientation and the well-known chroma-key technique to insert actors and props in a synthetic background. The techniques developed in this project will partially or completely suppress the constraints of chroma key. The results of the project will give the consortium a competitive edge on players from outside Europe, mainly from the US. Contact Point Mr. Pierre Pleven tel: +33.1.3463.7556 Symah Vision fax: +33.1.3463.7057 Rue Dewoitine 6 78140 Vélizy-Villacoublay FRANCE Participants Symah Vision Matra Systèmes et Information Radamec Broadcast Systems Digitale Videosysteme INRIA Mondial Promotions S.A. Start Date January 1997 Country F F UK D F GR Duration 24 months Role C P P P P P This page is located at www.cordis.lu/esprit/src/23483.htm It was last updated on 8 septembre 1998, and is maintained by Patrick.VanHove@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 86 23486 PHASE A Distributed Pharmaceutical Applications Server Keywords: High-Performance Computing and Networking Technical - Metacomputing, Parallel computing, Simulation, Internet Uses - Molecular modeling Industrial Applications - Drug design, biotechnology research The PHASE project aims at providing a central access point to powerful computing platforms running applications used in the pharmaceutical drug discovery process (e.g. MaxHom, Genequiz, Dragon and Msap). By linking WAN connected HPCN systems running these applications, a service is provided to industrial users who need high performance computing power for their specific tasks in drug design projects, but find the cost to be prohibitive, or have not decided yet to invest in the acquisition and management of the software and platforms. Building on previous projects, and on a national metacomputing initiative, PHASE combines important paralel bio-informatics applications for collaborative use in a distributed server environment. Based on the results and experiences gained in these projects, PHASE has the potential to demonstrate the feasibility and cost-effectiveness of HPCN in industrial key-applications in the field of pharmaceutical drug design, and in biotechnology research. Moreover the project intends to demonstrate that this type of service, accessible also through Internet, may be extended to other application codes, and other industrial sectors. Contact Point Mrs Anke Weber University Paderborn Paderborn Centre of Parallel Computing Fürstenallee 11 D-33102 Paderborn tel +49.5251.60.6331 fax +49.5251.60.6297 email: weber@uni-paderborn.de http://www.uni-paderborn.de/pc2/projects/phase Participants PC² Merck Novo EBI MRC-NIMR OML Country D D DK UK UK UK Start Date February 1997 Duration 24 months Role C P P P P A This page is located at www.cordis.lu/esprit/src/23486.htm It was last updated on 18 November 1998, and is maintained by Massimo.Luciolli@cec.be Octobre 1998 High Performance Computing and Networking / 87 23499 DYNAMITE Dynamic Task Migration Execution Environment Keywords: High Performance Computing and Networking, Dynamic Load Balancing User - Workstation Clusters, Parallel Job Management Industrial Applications - Crashworthiness, Occupant Safety Analysis, Metal Stamping Project Summary This project aims at the development of a DYNAMIc Task migration execution Environment (DYNAMITE) for the automatic work-load balancing of PVM tasks in workstation cluster environments with dynamically changing load. DYNAMITE will drastically lower the barrier to use high performance computing in industry and academia by optimally exploiting the distributed computing power of the locally installed workstations. It will allow industrial users to migrate to parallel program architectures without having to invest heavily in new hard- or software. With our combined expertise in modeling and simulation, parallel and distributed computing, task migration, scheduling and load balancing, we are in the unique situation to address the complex issues related to the efficient exploitation of workstation clusters for large scale simulations. The project is tackled along four mutually interacting layers: • task partitioning by ESI and UvA (for the PAM-CRASH code), • task load balancing by PC 2 (deciding when and where to migrate a task), • task migration by UvA (checkpointing, task transfer and restart), • industrial exploitation and dissemination by ESI and Genias. While the project initially targets at improving the execution speed of ESI's widely used car crash simulation code on workstation clusters, the resulting software environment is completely transparent to the application. Any existing PVM application can be linked to the DYNAMITE library without the need to change the source code. Contact Point Jan Clinckemaillie Engineering Systems International 20, rue Saarinen SILIC 270 F-94578 Rungis tel.: +33 1.49 78 28 00 fax: +33 1.46 87 72 02 email: jc@esi.fr http://www.esi.fr Participants Engineering Systems International Universiteit van Amsterdam Genias Benelux BV Paderborn Center for Parallel Computing Country Role F C NL P NL P D P Start Date September, 1997 Duration 2 years This page is located at www.cordis.lu/esprit/src/23499.htm It was last updated on 8 septembre 1998, and is maintained by khalil.Rouhana@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 88 23502 FITS Fortran Integrated Tool Set Keywords: High-Performance Computing and Networking, Simulation Technical : Reverse engineering, performance analysis, hybrid architectures, MPI Uses : Fortran, Code re-engineering, parallel code, toolset The goal of the project is to create an extensible integrated toolset supporting all phases of the creation of a modern HPC application, from existing code re-engineering to parallel code maintenance. This unique toolset on the market will reach the following strategic goal : • Bundle two major European tools dedicated for the support of FORTRAN applications : FORESYS from SIMULOG and VAMPIR from PALLAS and transfer technology from the prototype tools ANALYST from VCPC and IDA from the University of Southampton. • Integrate the existing tools in a seamless graphical framework providing easy to use graphical program representations transparently linked to textual views, support for emerging hybrid architecture, support for cache utilisation, MPI usage analysis The primary evolution criterium for FITS success will be the end-user acceptance and increase in productivity of code porting, optimization and maintenance activities. New modules which could subsequently be added to FITS would include HPF compilers and tools, as well as additional MPI-based tools. Contact Point Mr. Laurent HILL tel:+33.4.93 65 25 46 SIMULOG fax+33.4.93 65 25 57 Les Taissounières HB2 email: Laurent.Hill@sophia.inria.fr Route des Dolines http://www.vcpc.univie.ac.at/activities/projects/FITS 06560 Valbonne FRANCE Participants SIMULOG VCPC PALLAS INRIA BATTELLE Q.S.W TU. Dresden From 1.10.97 Start Date June 1997 Octobre 1998 Country F A D F D UK D Role C P P A A A A Duration 24 months High Performance Computing and Networking / 89 23516 WINPAR Windows Based Parallel Computing Keywords: High Performance Computing and Networking Technical - Application Development and Simulation Environment, Parallel Computing, Windows NT Uses - Training, Education and Development in Parallel Computing for Universities, Research and Industry Industrial Applications - Complex, Industrial and Scientific Applications Within the WINPAR project an integrated software development environment for parallel computing is developed, targeting personal computers interconnected by local area networks running Windows NT. The technical objectives of WINPAR are to provide a message passing layer including MPI and PVM, to provide a set of tools for code development, simulation, performance prediction graphical high-level debugging, monitoring and visualisation of parallel applications and a parallel computing tutorial. The commercial objectives are to offer an affordable parallel development environment for training and education at universities, research organizations and industry. Tools including AUGUR, MODARCH, TRAPPER, WPVM and WMPI are enhanced, integrated and ported to Windows NT. The WINPAR environment will be available for both, UNIX and Windows NT and share a common look-and-feel. WPVM and WMPI are offered as public domain on the Internet. The complete WI NPAR environment will be offered as a commercially available product offered by GENIAS. Already the first software component PaTENT MPI 4.0 has been released. Contact Point Wolfgang Gentzsch tel: +49 9401 92000 GENIAS software GmbH fax: +49 9401 920092 Erzgebirgstrasse 2 email: gentzsch@genias.de D-93073 Neutraubling http://www.genias.de/projects/winpar/index.html Participants GENIAS Software GMD KFKI-MSZKI SIMULOG University of Coimbra University of Miskolc University of Vienna Dash Associates TRANSVALOR Country D D H F P H A UK F Start Date January1997 Duration 25 months Role C P P P P P P A A This page is located at www.cordis.lu/esprit/src/23516.htm It was last updated on 10 November 1998, and is maintained by agnes.bradier@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 90 HPCN Preparatory Support and Transfer Activities (PST) 23713 TTN-T Technology Transfer Node Thuringia Keywords: HPCN, TTN, Embedded Systems; Machine Vision, QualityInspection Industrial Applications: Inspection Systems for Production and Manufacturing Processes; Optics, Image and Signal Processing, and other branches relevant to the New German Federal States. TTN-T is built on three non-profit organisations (Transferzentrum Mikroelektronik e.V. in Erfurt, Zentrum für Bild- und Signalverarbeitung e.V. in Ilmenau, and Gesellschaft für Fertigungstechnik und Entwicklung e.V. in Schmalkalden) with complementary tasks and skills in: Image processing and acquisition, production engineering for user support/demonstration, and transfer/project management for broker function. The consortium is located in Thuringia and will co-ordinate the European PST activities for HPCN application projects with the specific objective to integrate HPCN technologies for fast vision and "cognitive automation" into manufacturing and processing lines. The activities focuse mainly on real time defect detection in textures and structured componenents, 2D and 3D surface analysis for geometric objects and complex structures in (ultra-)fast processes, and quality inspection systems based on image processing in general. TTN-T will also promote R&D projects in HPCN embedded systems. Contact Point Dietmar Starke Transferzentrum Mikroelektronik e.V. In den Weiden 7 D-99099 Erfurt Tel: +49 361 4420664 Fax: +49 361 4420666 E-mail: tzm.erfurt@t-online.de URL: http://www.tzm.de Start Date Duration March 1997 24 months Octobre 1998 High Performance Computing and Networking / 91 HPCN Preparatory Support and Transfer Activities (PST) 23770 FINNOVA Novel Approaches to Finnish Industrial Computing Keywords: Simulation, High-performance Networking, CFD, Structural Analysis, Computational Chemistry, Numerical Methods, Visualisation, Network-based Multimedia, Multicast Networking, Code Optimization, Code Parallelization, Software Development, Bioinformatics, High Volume Databases Industrial Applications: Metal, Pulp and Paper, Energy, Electronics, Semiconductor, Chemical, Pharmaceutical Center for Scientific Computing (CSC) is the Finnish national center for high-performance computing and networking. It runs a large-scale facility for computational science and engineering and its customers come from universities, research institutes and industry. CSC is also responsible for the operations of the Finnish University and research network (FUNET). As a TTN, CSC specializes in high-end numerical simulations on parallel supercomputers using state-of-the art numerical methods and parallelization strategies. Our special expertise includes computational physics, chemistry and biology, CFD and structural analysis together with visualization and highperformance networking techniques. Contact Point Sirpa Kotila tel: +358 9 4572160 Center For Scientific Computing (CSC) fax: +358 9 4572302 e-mail: sirpa.kotila@csc.fi Tietotie 6 FIN - 02101 Espoo Start Date Duration March 1997 24 months Octobre 1998 High Performance Computing and Networking / 92 24003 HIPERTTN Hipercosme TTN Keywords: HPCN, TTN, Simulation, FEM, Early Fault Diagnosis, Quality Control Image Processing, Tolerance Analysis Industrial Applications: Construction Engineering, Turbines, Automotive, Aerospace, Textile, Machine-Tool Within the HIPERTTN project several assessments to end users with the demonstrators developed in the HIPERCOSME project will be performed in Pais Vasco, Comunidad Valenciana, MidiPyrénées, Attica and Porto regions. In Porto a demonstrator and awareness campaign will be developed. The assessments will be related to construction engineering, early fault diagnosis in gas turbines, quality control applying image processing and tolerance analysis in 3D mechanical assemblies. The main objective is to demonstrate, promote and introduce the use of HPCN techniques in SMEs solving their problems. Contact Point Jose Ignacio Barbero (Co-ordinator) Alfonso Longo LABEIN Cuesta de Olabeaga 164 E-8013 Bilbao Vicente Hernandez Universidad Politecnica de Valencia UPV - Camino De Vera Comunidad Valenciana E-46022 Valencia Pierre-Henri Cros CERFACS Avenue G. Coriolis 42 F-31057 Toulouse Anastasios Stamatis NTUA Iroon Polytechniou 9 GR-15710 Athens Armando Padilha FEUP Rua Dos Bragas P-4099 Porto Cedex Start Date April 1997 Tel: +34 4 4892400 Fax: +34 4 4892460 Email: jose@labein.es longo@labein.es Tel: +34 6 38 77356 Fax: +34 6 3877359 E-mail: vhernand@disc.upv.es Tel: +33 5 61 19 30 02 Fax: +33 5 61 19 30 30 E-mail: cros@cerfacs.fr Tel: +30 1 7721638 Fax: +30 1 7721658 E-mail: stamatis@saronicos.ltt.mech.ntua.gr Tel: +351 2 204 1839 Fax: +351 2 2087310 E-mail: padilha@garfield.fe.up.pt Duration 24 months This page is located at www.cordis.lu/esprit/src/24003.htm It was last updated on 8 septembre 1998, and is maintained by Max.Lemke@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 93 HPCN Preparatory Support and Transfer Activities (PST) 24396 ATTN Austrian Technology Transfer Node Keywords: HPCN, TTN, Simulation, River Level, Image Processing Industrial Applications: Casting, Waterway Management, Digital Film Restoration ATTN focuses on initiating, supporting and disseminating results of activities in simulation and design, data and information management, software tools for application development, and fast networking. Awareness campaigns will be carried out in areas of relevance to Austrian industry. Activities will spread the use of simulation applications, particularly for small business. Support will be given to the use of programming tools for application development. Further activities will include applying new solution methods, such as genetic algorithms and neural networks, and new application areas, particularly in multimedia and multimedia databases. Experience in using workstation clusters with ATM technology will be shared. Contact Point Barbara Chapman, Philippe Devillers VCPC Vienna Centre for Parallel Computing Liechtensteinstr 22 A-1090 Vienna Start Date July 1997 24 months Octobre 1998 Tel: +43 1 310 93 96 Fax: +43 1 310 93 96 13 E-mail: attn@vcpc.univie.ac.at Duration High Performance Computing and Networking / 94 HPCN Preparatory Support and Transfer Activities (PST) 24404 ENTICE European New Technology for Industrial and Commercial End-Users Keywords: HPCN, TTN, Optimisation, Visualisation, Radiotherapy, Data Warehousing, Inspection Industrial Applications: Architecture, Medicine, Manufacturing, Law The ENTICE TTN is led by Edinburgh Parallel Computing Centre. Its objective is to raise the awareness of HPCN technology both throughout the UK and across the rest of Europe. This will be strongly supported by a set of pilot projects primarily involving SMEs. The significance of HPCN to industry and commerce will be emphasised through the use of the excellent links already established between EPCC and local and regional Governmental and trade organisations. Initial activities, which span a wide range of areas, include industrial process optimisation, architectural visualisation, radiotherapy treatment planning, data warehousing and visualisation, and textile inspection. ENTICE has a broad range of HPCN technologies from PC-clusters to large supercomputers. These systems provide important demonstration platforms for the indsutry-focused dissemination and demonstration activities. Contact Point Mark Sawyer EPCC The University of Edinburgh James Clerk Maxwell Building Mayfield Road UK-EH9 3JZ Edinburgh Mark Parsons EPCC The University of Edinburgh James Clerk Maxwell Building Mayfield Road UK-EH9 3JZ Edinburgh Start Date April 1997 Octobre 1998 Tel: +44 131 650 5019 Fax: +44 131 650 6555 E-mail: m.sawyer@epcc.ed.ac.uk Tel: +44 131 650 5022 Fax: +44 131 650 6555 E-mail: m.parsons@epcc.ed.ac.uk Duration 24 months High Performance Computing and Networking / 95 HPCN Preparatory Support and Transfer Activities (PST) 24406 TTN NETWORK CO-ORDINATOR 2 Mechanism for Enabling HPCN Technology Transfer in Europe 2 Keywords:HPCN, TTN, Dissemination, Technology Transfer, Take-up, Stimulation, Cootdination, Awareness Creation, Promotion Industrial Applications: The HPCN TTN Network co-ordinators are providing a global infrastructure to support the activities of the TTNs. The HPCN TTN Co-ordinators aim at: • making the overall network of TTNs more coherent, effective and relevant to the outside world, • achieving a wide industrial impact among potential users of the HPCN technology, especially SMEs. The duties of the HPCN TTN Network Co-ordinators cover a wide range of activities facilitating co-operation and dissemination within the network of TTNs and ensuring successful dissemination and concertation of all PST activities. The HPCN TTN Network will support a professional awareness creation and dissemination campaign conducted at a local level by regional TTNs and co-ordinated at a global level by the HPCN TTN Network Co-ordinators. The campaign will be visible to a wide cross-section of European end-users through the use of a variety of media, including: the press, exhibitions and conferences, flyers and brochures, and multi-media, focusing on awareness creation and dissemination among new end-user groups. Achievement of these objectives will produce the added value of consistent and mutually compatible TTN results thus increasing the impact of the programme. Contact Point Eric Papon, Carlos Triay ARTTIC 5, Avenue de Verdun F-94204 Ivry Sur Seine CÈdex Start Date March 1997 Octobre 1998 Tel: + 33 1 45 15 24 50 Fax: + 33 1 45 15 24 60 E-mail: hpcn-ttn@list.arttic.com URL: http://www.hpcn-ttn.org Duration 30 months High Performance Computing and Networking / 96 HPCN Preparatory Support and Transfer Activities (PST) 24410 TTN NETWORK CO-ORDINATOR 1 Mechanism for Enabling HPCN Technology Transfer in Europe 1 Keywords: HPCN, TTN, Dissemination, Technology Transfer, Take-up, Stimulation, Co-ordination, Awareness Creation, Promotion As an HPCN TTN Network Co-ordinator, Smith System Engineering is responsible for identifying synergy between TTNs and opportunity and exploitation potential across the entire TTN Network. It facilitates co-operation and dissemination within the network of TTNs ensuring successful dissemination and concertation of all PST activities. The Network supports a professional awareness creation and dissemination campaign conducted at a local level by regional TTNs and co-ordinated at a global level by the Network Co-ordinators. The campaign will be visible to a wide cross-section of European end-users through the use of a variety of media, including: the press, exhibitions and conferences, flyers and brochures, and multi-media, focusing on awareness creation and dissemination among new end-user groups. The together with its co-ordinating partner ARTTIC, Smith hopes to accelerate the take-up of the technology through supporting the successful demonstration of cost-effective HPCN solutions. In keeping with the primary goals of the TTN Network, this work focuses on helping European industry (and in particular SMEs) to understand the cost-benefits and best practice in using HPC and HPN. Contact Point Daron Green Smith System Engineering Ltd Surrey Research Park UK-GU2 5YP Guildford Start Date March 1997 Octobre 1998 Tel: +44 1483 442000 Fax: +44 1483 442304 E-mail: dggreen@smithsys.co.uk URL: http://www.hpcn-ttn.org Duration 24 months High Performance Computing and Networking / 97 HPCN Preparatory Support and Transfer Activities (PST) 24414 NOTSOMAD Nodo di Transferimiento Technologico a Sistemi di Operatori del Manifatturiero e dei Servizi Keywords: HPCN, TTN, Simulation, Quality Inspection Industrial Applications: PCB Quality Control The NOTSOMAD TTN will addresse primarily the traditional sectors of the Italian industry, represented by small and medium enterprises of textile, mechanics, ceramics, food, wood, etc. Its aim is to demonstrate by example that competitiveness can be achieved by using very innovative and powerful simulation tools and quality inspection systems based on HPCN technology. A special attention will be devoted to making newcomers aware of the great benefits achievable by adopting new HPCN-based Decision Support Systems in market, commercial and financial fields. The NOTSOMAD TTN hopes to show that often unexpectedly high performance levels can be achieved through properly clustering workstations and even personal computers. Such systems often represent very low-cost solutions to small businesses. Contact Point Mafalda Valentini,Massimo Busuoli (Co-ordinator) ENEA Inn-Diff Via Don Fiammelli, 2 I-40129 Bologna Dr Mario Lanzarini, Guiseppe Paruolo Cineca Via Magnanelli 6/3 I-40033 Casalecchio Di Reno Start Date Duration March 1997 24 months Octobre 1998 Tel: +39 51 609 8203 / 8178 Fax: +39 51 609 8084 E-mail: valentini@notsomad.org Tel: +39 51 6171448 Fax: +39 51 6592581 E-mail: lanzarini@cineca.it High Performance Computing and Networking / 98 HPCN Preparatory Support and Transfer Activities (PST) 2442 TTNMV-SF Technology Transfer Node of Machine Vision in Finland Keywords: HPCN, TNN, Embedded Systems and Networking; Machine Vision, Colour Vision, Distributed Computing Industrial Applications: Machine Vision Applications in Different Industries; Process Automation (Pulp and Paper, Graphic Arts, Burning Process, Condition Monitoring, Food and Beverage, Mineral Processing...) , Robot Vision (Pattern Recognition, 3D Positioning and Measuring...), 3D Inspection and Modelling, Quality Control (Surface QC and Grading) TTNMV-SF has a focused mission to exploit HPCN technology in the area of machine vision. In addition, it hopes to stimulate the transfer of technology and expertise across both Finland and the rest of Europe. The TTNMV-SF activities will bring machine vision technology to new industries and especially to SMEs. Its Demonstrations and Best Practice actions will convince the European wide industry of the viability of those young machine vision providers and their new products. Technology experts in research organisations can support providers of machine vision solutions in upgrading their system capabilities with HPCN techniques. This allows component suppliers to offer more cost-effective/new products and services. Contact Point Antti Soini Tel: +358 9 456 6175 Finnish Automation Support Ltd Fax: +358 9 456 6752 E-mail: antti.soini@innopoli.fi Tekniikantie 12 Fin-02150 Espoo Mailing address: P.O. Box 1301 FIN-02044 VTT Start Date Duration March 1997 24 months Octobre 1998 High Performance Computing and Networking / 99 HPCN Preparatory Support and Transfer Activities (PST) 24448 CAPRICE Centre for Accessing HPCN Best Practice Information, Technology and Application Environments Keywords: HPCN, TTN, Simulation, Embedded Systems, Multi-site Applications, Development Environments Industrial Applications: electronic industry, transport logistics, aerospace, automotive The technology transfer node CAPRICE provides access to information, technology and application environments for high performance computing and networking. In this context, the utilisation of workstation or personal computer clusters for high performance computing results in high performance computing and networking representing more and more two sides of the same medal. Therefore, CAPRICE explicitly addresses the combination of high performance computing and networking. The focus of CAPRICE is to support information exchange on high performance computing and networking best practice for small and medium size enterprises working in the technical sector. This includes networked multi-site applications for simulation and embedded systems as well as technical information managing and decision support systems. In this framework special attention is paid to the needs and constraints of small and medium size high technology enterprises. Therefore, not only know-how and solutions suitable for small and medium size enterprises are covered, but also special services required in order to enable small and medium size enterprises to apply for funding of their preparatory, first user, demonstration or best practice are offered. Typical concepts and applications addressed in the projects and activities associated to the Technology Transfer Node CAPRICE are • telecollaboration in virtual enterprises consisting of several distributed small and medium sized enterprises, • tele-engineering and telesimulation in the framework of high technology development efforts, • remote and distributed software engineering in virtual teams. Further promising telecooperation and telecollaboration applications are envisaged in the automotive, construction, medical, environment and transport, pharmaceutical/chemical and finance, insurance and retail domain. Contact Point Gabriele Pawlitzek, Werner Rohmann Deutsche Forschungsanstalt f¸r Luft- und Raumfahrt (DLR) Central Data Processing Division Lilienthalplatz 7 D-38108 Braunschweig Start Date April 1997 Octobre 1998 Tel: +49 531 295 2767 (G. Pawlitzek) Tel: +49 531 295 2975 (W. Rohmann) Fax: +49 531 295 2880 E-mail: etc.caprice@dlr.de URL: http://www.eurofast.de/ttn/ Duration 24 months High Performance Computing and Networking / 100 HPCN Preparatory Support and Transfer Activities (PST) 24462 TTNatGMD Technology Transfer Node at GMD Keywords:HPCN, TTN, Simulation Industrial Applications:Surgery, Chemical, Automotive, Forging TTNatGMD is focused on demonstrating the benefits of HPCN to those previously unaware of them. This will be achieved through active awareness creation campaigns and through actively encouraging and helping newcomers to use HPCN techniques. Initial activities focus on the automotive industries, the Chemical industry, Forging processes, Medicine and PC based simulation. Contact Point Ottmar Krâmer-Fuhrmann GMD-Forschungszentrum Informationstechnik GmbH SCAI-Institute For Algorithms Schloss Birlinghoven D-53754 Sankt Augustin Start Date March 1997 Octobre 1998 Tel: +49 2241 14 2202 Fax: +49 2241 14 2386 E-mail: ttn@gmd.de Duration 24 months High Performance Computing and Networking / 101 Start DAteHPCN Preparatory Support and Transfer Activities (PST) 24484 INNO An Inspection Technology Transfer Node Keywords: Machine Vision, Distributed Visual Inspection, Camera Imaging, Image Processing, Pattern Recognition, Optical Character Recognition (OCR), Classification, Sorting, Parallel Processing, Robotics, Distributed Processing, Networking, Artificial Neural Networks Industrial Applications: Textile, Electronics, Manufacturing, Packaging, Material Placement, Material Soldering, Cutting and Sewing, Painting, Quality Control, Audio Visual Production, Digital Video Production, Pharmaceutics, Stationery and Paper Production, Content Creation, Decision Making and Support, Information Management The main focus of INNO is HPCN-enabled machine vision applications for quality control in production lines, though not exclusively. INNO aims at developing a centre of expertise that has the know-how for the introduction of valueadded services directly to the industrial sector and the production line. These services include identification of particular needs and customisation of the available solutions, setting up a prototype and supervising the operation, measuring the performance against prespecified goals and incorporation of the complete process experience into the value added chain. In turn, these results will further boost INNO's ability and flexibility to offer a concise, time critical, professional knowledge to industrial partners. INNO activities are targeted to the current level of HPCN awareness and adoption in the user organisations through different means. Assessments refer to the establishment of user requirements, feasibility studies or early prototyping and are usually of specific applicability. Demonstrations relate to the adoption by transition from traditional to HPCN technology, while Best Practices refer to re-engineering of HPCN applications and to transfer and ensuring take-up in new user environments. The evaluation of the experience gained within the activities will help produce a set of replicable techniques which will be used for subsequent activities. Activity milestones comprise: (1) user requirements definition, (2) application architecture design, (3) market re search for tools and applications, (4) feasibility study and cost/benefit analysis and (5) early prototype implementation. Initial activities relate to quality control in the textile sector, electronics industry, audio visual production and integration of HPCN-based neural network models to the provision of intelligent decision support. Contact Point Apostolos Meliones National Technical University of Athens Dept. of Electrical and Computer Engineering Computer Science Division 9 Heroon Politechniou str. GR-15773 Zografou-Athens Start Date March 1997 Octobre 1998 Tel: +30 1 772 1478 / 2076 Fax: +30 1 772 2534 / 2077 E-mail: meliones@cs.ntua.gr URL: http://www2.y-net.gr/inno/ Duration 24 months High Performance Computing and Networking / 102 HPCN Preparatory Support and Transfer Activities (PST) 24617 DANHIT Danish HPCN Technology Transfer Node Keywords: HPCN, TTN, CFD, Manufacturing Technology, Bioinformatics, FEM Simulation Industrial Applications: Manufacturing, Medical, Pharmaceutical, Food, Environmental agencies, Transport The DANHIT TTN aims to support and promote the take-up of HPCN applications and technologies both within its local region and in the broader Network context. DANHIT will initially concentrate on the following main themes, selected from areas in which the Danish community is particularly strong: Bioinformatics, in close collaboration with the BIOTITAN TTN, e.g. genetic analysis of selected traits in animal populations. Flexible manufacturing technology with a focus on advanced industrial robots for short or complex productions runs. Environmental studies with a focus on airborn pollution models. The medical sector, e.g. anaesthesia simulation and design of human prostheses. Aerodynamic analysis of airflow impact, e.g. aerolastic bridge stability simulation and design of wind rotor blades. Contact Point Jorgen Moth UNI-C The Danish Computing Centre for DTU, Bldg. 304 DK-2800 Lyngby Start Date April 1997 Octobre 1998 Tel: +45 3587 8963 Fax: +45 3587 8990 E-mail: jorgen.moth@uni-c.dk Duration 24 months High Performance Computing and Networking / 103 HPCN Preparatory Support and Transfer Activities (PST) 24620 TETRAPC Technology Transfer in Parallel Computing Keywords: HPCN, TTN, Simulation, Signal Processing, Image Processing, Pattern Recognition, 3D Reconstruction, Quality Inspection using Vision Technologies, Artificial Neural Networks, CIM, Information Management Industrial Applications: Textile/Clothing, Marble, Mechanical/Iron And Steel, Hides Tanning, Furniture (Limited To Process Control), Electro-Medical, Credit, Public Administration, Water/Gas/Electricity Distribution, Cultural Heritage. TETRAPC aims at unifying the existing know-how and expertise in HPCN technologies in Italy at local, national and transnational level and supporting enterprises to take-up these technologies. TETRAPCÌs activities will mainly focus on SMEs, aiming at the following objectives: adapting the HPCN supply to the needs of enterprises; promotion of HPCN technologies; acting as a focal point between HPCN users and technology suppliers in order to improve enterprises competitiveness both at national and transnational level, across the TTNs and IRCs network. Contact Point Cinzia Giachetti & Alberto Ciampa Conzorzio Pisa Ricerche Innovation And Technology Transfer Piazza A. D Ancona 1 I-56127 Pisa Start Date March 1997 Octobre 1998 Tel.: +39 50 906260 Fax: +39 50 540056 E-mail: tetrapc@cpr.it.ciampa@ipifidpt.difi.unipi.it Duration 24 months High Performance Computing and Networking / 104 HPCN Preparatory Support and Transfer Activities (PST) 24621 ICETACT Irish Centre for Transfer of Advanced Computing Technology Keywords:HPCN, TTN, Remote Sensing Industrial Applications: Exploration ICeTACT aims to execute a programme to bring HPCN to the primarily SME sector in the North and the South of Ireland and to disseminate the results obtained across Europe through the network of TTNs. The consortium forming ICeTACT is committed to continuing the work that it will execute with European support beyond the funding period of the project. The initial activities will assess the uses of HPCN in the remote sensing and exploration support services industry. Contact Point Eoin O'Neill Uni. Dublin Trinity College Innovation Services College St 2 IRL-Dublin Prof. Ron Perrot QueenÌs University Department Of Computer Science UK-BT7 1NN Belfast Mr Declan Martin Dublin Chamber Of Commerce Clare Street 7 IRL-Dublin 2 Start Date April 1997 Octobre 1998 Tel: +353 1 608 1427 Fax: +353 1 679 8039 E-mail: Eponeill@Tcd.Ie Tel: +44 1232 335463 Fax: +44 1232 683890 E-mail: r.perrott@qub.ac.uk Tel: +353 1 6614111 Fax: +353 1 676 6043 E-mail: declan@dubchamber.ie Duration 24 months High Performance Computing and Networking / 105 HPCN Preparatory Support and Transfer Activities (PST) 24722 BIOTITAN Transfer of Bioinformatics into European Industry Keywords: HPCN, TTN, Bioinformatics Industrial Applications: Animal Breeding, Visualisation of large Databases The objective of the BIOTITAN TTN is to stimulate the uptake of HPCN by the biotechnology, biomedical, agricultural and pharmaceutical industries in Europe. This is accomplished through PST activities coordinated with all members of the HPCN TTN Network. BIOTITAN is comprised of three partners: the European Bioinformatics Institute (EBI, coordinating partner), the Danish Computing Centre for Research and Education (UNI − C)and the Consortium for Supercomputing Applications for University and Research (CASPUR). The BIOTITAN partners bring together resources and expertise in bioinformatics, molecular biology, computational chemistry, parallel and high performance computing, industrial computer applications, network infrastructure and computer support services. BIOTITAN’s mission is to apply this advanced HPCN technology to the benefit of commercial end-users in the target industrial sectors, making them more competitive in the global market. Contact Point David Sanders (Co-ordinator) Tel: +44 1223494441 The European Bioinformatics Institute (EBI) Fax: +44 1223494470 Wellcome Trust Genome Campus E-mail: biotitan@ebi.ac.uk Hinxton UK-CB10 1SD Cambridge Jorgen Moth Tel: +45 358 78889 Uni-C Fax: +45 35878990 Scientific Computing E-mail: danhit@uni-c.dk DTU, Building 304 DK-2800 Lyngby Nico Sanna, Romano Bizzarri Tel: +39 6 49913 997/539 CASPUR Fax: +39 6 4957083 P. le Aldo Moro 5 E-mail: sanna@caspur.it I-00185 Roma Start Date Duration April 1997 24 months Octobre 1998 High Performance Computing and Networking / 106 HPCN Preparatory Support and Transfer Activities (PST) 24724 PROHPC Promotion of High Performance Computing Keywords: HPCN, TNN, Simulation, Medical Imaging, Actuarial Services Industrial Applications: Active Noise Control, Drill Bit Design, Insurance Pricing, Medical Imaging, Acoustic Design The main objective of the ProHPC TTN is to increase both the use of HPCN technologies and relevant how-how for enhancing industrial competitiveness, particularly towards the many SMEs and industrial French companies, who have little access to such technologies. With four partners covering several regions, this TTN has regional actions linked at a national level. The TTN will re-use the results and porting experience of the EuroTOPS Eureka project, other results from European and national projects, such as the European Porting Projects, as well as the industrial and technical expertise of its partners. The individual activities cover a large spectrum of technologies and needs: Awareness Campaign and Assessments of Rhone-Alpes region SMEs, Acoustics, Drilling Tool Simulation, Financial Services and Medical Imaging. Contact Point Karine Van Heumen, Yves Robert (co-ordinator) Ecole Normale Supérieure de Lyon - LIP 46, allée d'Italie F-69364 Lyon Cedex 07 Tel: + 33 (0)4 72 72 84 58 Fax: + 33 (0)4 72 72 80 80 E-mail: ttnpoc@ens-lyon.fr Thierry Priol INRIA Campus de Beaulieu F-35042 Rennes Tel: +33 (0)2 99 84 72 10 Fax: +33 (0)2 99 84 71 71 E-mail: Thierry.Priol@irisa.fr Start Date March 1997 Duration 24 months Octobre 1998 High Performance Computing and Networking / 107 HPCN Preparatory Support and Transfer Activities (PST) 24727 PDC TTN Swedish HPCN Technology Transfer Nodes at PDC Keywords: HPCN, TTN, Graphics, Video Animation, simulation, data management Industrial Applications: All sectors relevant for the region PDCTTN will contribute to the overall objective of the HPCN initiative by facilitating technology transfer between, on one hand, Swedish end-users and software developers and, on the other hand, initiate contacts with similar groups in other European regions. PDCTTN will organise workshops in Sweden and possibly in other Northern European countries to promote user awareness of HPCN technology. Initial activities relate to the parallelisation of video graphics effects, simulation and data management systems. Contact Point Marina Backer Skaar, M.Sc. TTN co-ordinator / PDCTTN Parallelldatorcentrum (PDC) Royal Institute of Technology S-100 44 Stockholm Start Date April 1997 Octobre 1998 Tel: +46 8 790 6923 Fax: +46 8 24 77 84 E-mail: pdcttn@pdc.kth.se URL: http://www.pdc.kth.se/pdcttn Duration 24 months High Performance Computing and Networking / 108 HPCN Preparatory Support and Transfer Activities (PST) 24740 AVS-TTN Advanced Video Surveillance Technology Transfer Node Keywords: HPCN, TTN, Video Surveillance, Security Industrial Applications: Theft Prevention, Vandalism Prevention, Staff Protection, Customer Protection AVS-TTN focuses on advanced video surveillance (AVS) techniques and applications, i.e. applications where video-cameras and image processing/understanding techniques, complemented, as required by the application, with other dedicated or integrated sensors, are used to detect "abnormal" events, which can then be recorded, used for automated control or forwarded to a human operator for final decision. Initial applications include using AVS techniques for the prevention of vandalism, theft and protection of employees and customers. Contact Points Philippe Cornez (Co-ordinator) CRIF - WTCM Automatisation et Gestion Industrielle Av. F. Roosevelt 50 CP106 - P4 B-1050 Bruxelles Tel: +32 2 650 39 88 Fax: +32 2 646 25 69 E-mail: cop@crif.ulb.ac.be Massimiliano Peri Advanced Engineering Technology Viale Mosto 4 1 I-16146 Genova Tel: +39 10 3774810 Fax: +39 10 3776518 E-mail: aetnet@mbox.vol.it Start Date March 1997 Duration 24 months Octobre 1998 High Performance Computing and Networking / 109 HPCN Preparatory Support and Transfer Activities (PST) 24742 DUTCH-TTN Dutch HPCN Technology Transfer Node Keywords: HPCN, TTN, Databases, Embedded Systems, Simulation, CFD, Visualisation, Metacomputing, Virtual Reality Industrial Applications: The DUTCH-TTN aims to deliver "ready-to-use" HPCN technology in a the standard business environment to strengthen the economic structure of the Netherlands. The activities of the DUTCH-TTN consist of an active brokering role for the different parties providing and requiring HPCN knowledge, applications and facilities. The DUTCH-TTN will actively contribute to a strong and internationally oriented market for HPCN-knowledge and applications in order to obtain sustainable co-operation between the public knowledge infrastructure and the private sector, especially SMEs. Current projects are involved in a wide range of sectors including: banking, aerospace, dredging, shipbuilding, chemistry, electrotechnics, food, cosmetics, pharmacy, civic engineering, environment technology, physical planning, traffic management, multi media, entertainment, design and training/learning and many others. Contact Point Jan Willem Tellegen Stichting HPCN Stadsring, 187 NL-3817 BA Amersfoort Start Date May 1997 Octobre 1998 Tel: +31 334 754 001 Fax: +31 33 472 2579 E-mail: BureauHPCN@MxI.nl Duration 24 months High Performance Computing and Networking / 110 HPCN Preparatory Support and Transfer Activities (PST) 24757 CEPBA-TTN CEPBA Technology Transfer Nodes Keywords: HPCN, TTN, Simulation, Information Management and Decision Support, CFD, CMD, Optimisation, High Volume Databases, Data Mining Industrial Applications: Fan Noise Reduction, Automotive, Packaging, Preventative Maintenance The CEPBA-TTN project aims at promoting the use of HPCN technology in industry by disseminating and stimulating the take up of the results of projects in the area of HPCN. The CEPBA TTN will be oriented towards the promotion of HPCN in industry, and specially in SMEs and newcomers. As TTN it aims to promote HPCN at large both in terms of end user industrial sector and technology (programming model, platform). Door to door marketing and close personal contact are the basis of the strategic approach of CEPBA to the promotion of the HPC technology. The TTN will focus in promoting regional activity and linking it to other European activities. Spain is its primary target area. Contact Point Jesus Labarta, Judit Gimenez Tel: +34 3 401 6987 / 7178 / 6986 UPC (Politecnical University of Catalunya) Fax: +34 3 401 7055 CEPBA E-mail: cepba-ttn@cepba.upc.es C/ Jordi Girona 1-3, Modulo D6 E-08034 Barcelona Start Date Duration March 1997 24 months Octobre 1998 High Performance Computing and Networking / 111 HPCN Preparatory Support and Transfer Activities (PST) 24758 ESCALATE Exploitation of Synergies and Co-operative Actions Leading to Accelerated Take-up and Effectiveness Keywords: HPCN, TTN, Simulation, Modelling, Information Management, Data Mining, Knowledge Extraction, High-Performance Databases, Knowledge Discovery and Structural Modelling Using Finite Element Analysis Industrial Applications: Retailing, Composite Materials, Acoustic Design, Automotive, OffShore, Loud Speaker Design, Noise Reduction, Aerospace, Retail, Structural Engineering Hi-Fi ESCALATE will promote and enhance the success and competitiveness of European industry through the increased uptake of HPCN technology. This will be enabled through demonstration of the business benefits of the technology and by the identification and exploitation of synergy between all ESCALATE projects, between ESCALATE and other TTNs and other initiatives. ESCALATE will utilise the Parallel Applications Centre's underlying application deployment and information dissemination skills and track record of successful partnering with European industry to support activities in the areas of simulation and modelling, information management, embedded systems and multimedia. Contact Point Paul Gordon Parallel Application Centre (PAC) University of Southampton 2 Venture Road Chilworth UK-SO16 7NP Southampton Start Date April 1997 Octobre 1998 Tel: +44 1703 760834 Fax: +44 1703 760833 E-mail: plg@pac.soton.ac.uk Duration 24 months High Performance Computing and Networking / 112 24897 HCHLOUSO Hydrocarbon and Chemical Logistics Optimization under Uncertainty Keywords: High Performance Computing & Networking, Decision Support, Optimisation Technical: Stochastic Optimisation Uses: Hydrocarbons, Chemicals Industrial Applications: Supply, Transformation and Distribution Logistics Scheduling The ability to solve large (in terms of the number of decision variables) and stochastic (in terms of parameters whose values cannot be controlled by the decision maker and are uncertain) industrial problems in practice depends on the sophistication of the tools and techniques available. The project will remedy the inadequacies of the tools currently available by developing flexible software for Supply, Transformation and Distribution (STD) logistics scheduling under uncertainty with access to new, more powerful methods capable of solving problems currently considered intractable. Specifically, the project aims to confer the ability to solve vital revenue critical problems for the industrial partners, namely, STD logistics scheduling under uncertainty for the Hydrocarbon and Chemical sector. The project addresses improvements in both quality (better scheduling, lower costs) and efficiency (shorter response times) in STD scheduling through parallel computing algorithm implementations. This will have a direct bearing on the industrial partners’ competitiveness world-wide as they face the challenges of deregulation and market globalization; Furthermore, the software developments which result will be of benefit to other industrial end-users beyond the consortium. HPC offers the means to obtain solutions very quickly, which can mean the difference between obtaining a practical solution and one which is too late for industrial purposes or is not the best solution that can be provided. The project will synthesise existing technology in the following domains: • high level model generation, • automatic uncertain data scenario generation, • optimization through recursion to deal with uncertainty and very large-scale problems, and • advanced data structures to benefit from the opportunity to use parallel computing Contact Point Laureano Escudero UITESA Avda. de Burgos 8B E-28036 Madrid Participants UITESA Agip Petroli CLH Dash Cambridge University IASI Start Date September 97 Octobre 1998 tel: +34 1 383 31 80 fax: +34 1 383 33 11 email: leb@uitesa.es Country E I E UK UK Role C P P P A I A Duration 30 months High Performance Computing and Networking / 113 24900 HPCN-SCAT HPCN-Stochastic Correlation of Analysis and Test Keywords: High-Performance Computing and Networking Technical - Monte Carlo Simulation Uses - Experiment validation Industrial Applications - Aerospace, Automotive, Railway The goal of the project is to advocate the development of a new HPCN-based methodology of numerical model validation that would be based on statistical correlation with experimental results. The correlation approach shall rely upon Monte Carlo techniques, already available within the framework of the PROMENVIR project, and shall enable a high level of genericity in terms of usage of numerical simulation codes. Performance of the statistical correlation and validation methodologies shall be evaluated using existing test data from the Aerospace, Automotive and Railway industry and contrasted with classical deterministic techniques. Contact Point Mr. V. Gomez Molinero tel +34-1-586 3796 CASA - Division Espacio fax +34-1-747 4799 Avda. Aragon 404 email: vgomez@casa-de.es E - 28022 Madrid http: to be announced Participants CASA Envision Intespace Italdesign ICA CIMNE Blue Engineering Country E E F I D E I Start Date September 1997 Duration 24 months Octobre 1998 Role C A P P A P P High Performance Computing and Networking / 114 24903 FLASH HPCN Tools for Enhanced Hydrodynamic Design of Fast Ships on Parallel Computing Platforms Keywords: -Hydrodynamic design, Computational Fluid Dynamics, Computational Structural Mechanics Technical - Parallel computing, Coupled CFD and CSM Finite Element system Uses - Light weight fast ships design Industrial Applications The overall goals of the project are: 1. to provide the fast ship construction community with a new HPCN system for efficient exploitation in the hydrodynamic design and optimisation of hull shapes and structures using heterogeneous parallel computing platforms composed of clusters of workstations and PC’s. 2. to demonstrate and assess the developed HPCN system in selected large-scale multi-physics computations typical in ship hull design involving coupling between hydrodynamics and structural mechanics problems. 3. to promote the HPCN system and disseminate the results of the demonstration actions among naval architecture industries by means of product marketing and organisation of workshops. 4. to raise interest for HPCN technology in the European ship construction industry by providing a user friendly parallel application software for both design and optimisation of fast and conventional ferries for passenger and cargo transportation which meets industrial needs. More specifically, the project will focus on development and integration in a parallel computing environment of existing finite elements based simulation software for CFD (Computational Fluid Dynamics) and CSM (Computational Structural Mechanics) to be used as efficient tools for solving relevant problems of the naval architecture industry Contact Point Mr. Antonio Pérez de Lucas Tel +34 1 335.85.11 BAZAN Fax +34 1 335.86.32 Castellana, 55 e-mail: aplucas@enbazan.es E-28046 Madrid Participants Country Role BAZAN ENVC BEC SP Technologies NAUTATEC CIMNE INESC Start Date October 1997 Octobre 1998 ES PT FR UK ES EE PT Duration 30 months C P P P P P P High Performance Computing and Networking / 115 24907 COLOMBO Parallel COmputers improve cLean up of sOils by Modelling BiOremediation Keywords: High Performance Computing and Networking, bioremediation. Technical - Mathematical simulation modelling, Cellular Automata, Polluted soils bioremediation. Uses - Project of Polluted soil remediation interventions, simulation modelling SW. Industrial Applications - Environmental sciences: project and realisation of soil remediation interventions, SW products: general SW environment for Cellular Automata modelling. Objectives The main objective of the project is the application of parallel computing to the simulation of the bioremediation of contaminated soils. In situ bioremediation is based on the use of bacteria to degrade the contaminant directly in the polluted soils, and can be seen as a way to improve and accelerate natural decontamination processes. As it is usually based upon the use of indigenous bacterial strains, it is environmentally safe and less expensive than other decontamination techniques. In order to predict the outcome of field scale operations from laboratory or pilot plant data and to reliably evaluate the times and the costs of the intervention, it is necessary to resort to mathematical models which describe the time evolution of the relevant variables during bioremediation. Since many physical, chemical and biological phenomena must be taken into account, these models require a large computing power which can be provided by parallel computers in a cost effective way. The most important objectives of the COLOMBO are: - the development and testing of cellular automata models, that could simulate a wider class of phenomena that occur in typical bioremediation interventions (three phase flow, pore clogging, interactions among bacterial populations); - the pilot plant testing of the model on a wide set of soil and contaminant types; - the model testing on a real field scale (not only in the pilot plant); - the development of a portable software environment for parallel MIMD machines, using the MPI standard; - the development of a "state of the art" tool for on line visualisation and interaction with the simulation; - the development of a software system which describes the spatial geological features of the site. The Project will build on the experience and on the results of the previous Esprit HPCN Project named CABOTO. Contact Point Massimo Andretta Tel. : +39.544.530016 Montecatini - Centro Ricerche Ambientali Fax : +39.544.538583 (Environmental Research Centre) email: mandretta@cramont.it Via Ciro Menotti, 48 I-48023 Marina di Ravenna (RA). Italy Participants Country Role CRA I C Umweltschutz Nord D P ENEA I P EPCC UK P QSW UK P Ironside Farrar Ltd. UK A CNR-ISI I A UNICAL I A Start date Duration January 1998 24 months Octobre 1998 High Performance Computing and Networking / 116 24916 VICAR Video Indexing, Classification, Annotation and Retrieval Keywords: High Performance Computing and Networking, Digital Video Technical - Image Analysis, Adaptive Pattern Recognition, Semantic Search Uses - Broadcasting, Movie Productions and Agencies, Security Industrial Applications - Automatic Generation of Semantic Indexes, Associative Search The overall aim of the project is to develop a system which includes each of the functions required to index & search videos. The primary focus is the professional video archive market, for which a pilot system will be developed. The requirements for adapting the software to smaller scale operations (consumer market) will be taken into account. The project will install a pilot version at the end user sites to measure user acceptance. The end users will create indexes of a high percentage of their digital video material and provide them to a central site, which will maintain them on an HPC system and provide a global search service. A WWW interface will enable access to the system. Quality and robustness of the search results is a major technical goal. The project will develop a software system for the automatic generation of semantic indexes for video and for associative search in the index structures thus formed. This environment will function as a learning system that is trained on reference material (video with content descriptions) to construct relevant high dimensional feature representations. The product Video Explorer of the project is a semantic indexing and retrieval engine for video databases. The project will develop software products for the creation of video indexes and for searching within such an index. It will also build up a prototype service which makes use of this software and the parallel computer at VECPC to enable a wider-scale search and retrieval of centrally stored index material. Although the service will begin with content provided primarily by the end users in this project, it is believed that in the longer term there is a large potential market both on the supply as well as on the demand side. A new company will be formed in order to fully exploit the results of the project. Contact Point Heimo Müller Joanneum Research Steyrergasse 17 A-8010 Graz Participants Joanneum Research Österreichischer Rundfunk Südwestfunk, Baden-Baden Sentient Machine Research Sveriges Television AB Faculteit der letteren Vrije Universiteit VECPC Start Date July 1997 Octobre 1998 tel: +43 316 876 242 fax: +43 316 876 191 email: Heimo.Mueller@joanneum.ac.at Country A A D NL S NL Role C P A P A A A A Duration 18 months High Performance Computing and Networking / 117 24949 IMPACT Inverse Methods for wave Propagation AppliCations in Time domain Keywords: High Performance Computing and Networking Technical: Parallel computing, simulation solver, inverse method Uses: Electromagnetism, acoustics, vibro-acoustics, elastodynamics, dielectric material Industrial Applications: Aircraft, satellites, medical equipment, geophysics The analyses of the response of a structure excited with a radiated signal concern several industrial applications and existing needs in: • electromagnetism: determining material properties of an object or its position and shape, • acoustics: analysing an object, determining its properties following an obtained response and determining the properties needed to obtain the desired response, • elastodynamics: analysing the density and composition of a medium. The objective of the IMPACT project is to address these needs by developing a parallel implementation of an inverse scattering method in the time domain. The method will be used to determine unknown physical properties of an object that are not available from direct measurements, difficult to obtain or to specify (design). The method will be applied to derive three inverse codes from three existing parallel solvers. The method and the inverse codes will be validated by three industrial test cases for electromagnetism (dielectric characterisation of a multi-layer structure) and for acoustics (improvement of the inside level noise for satellites and of the vibration levels at unit interfaces, and improvement of the focalisation of an ultrasound imaging medical equipment). Achievement of the project objectives will provide the following results: • The IMPACT inverse scattering method, with guidelines and generic software components to be used to derive an inverse code from an existing solver and to handle the inverse computations. This method will be further developed after the project as part of the exploitation with training modules and a service offer. With the project results, these additional developments will make a full product that will be marketed at the European level. • Three codes implementing the IMPACT inverse method: • the code from CRS4, a time domain acoustic volumic code, to be used in an Esaote ultrasound imaging medical equipment, • the code from KTH, a time domain electromagnetic code, to be used to perform electromagnetic analysis and optimisation, • the code from Aerospatiale, a time domain acoustic surface elements code, to be used for noise reduction in transport and spatial systems. Contact Point Mr Guillaume Alléon tel.: +33 1 46 97 36 83 Aerospatiale Centre Commun de fax: + 33 1 46 97 30 08 Recherches Louis Blériot 12 Rue Pasteur email: guillaume.alleon@siege.aerospatiale.fr F-92152 Suresnes Cedex http://www.arttic.com/projects/IMPACT/ Octobre 1998 High Performance Computing and Networking / 118 Participants Aerospatial CCR KTH ESB QSW CRS4 ESAOTE Start Date 1 November 1997 Country F S S UK I I Role C P P P P P Duration 24 months This page is located at www.cordis.lu/esprit/src/24949.htm It was last updated on 1 July 1998, and is maintained by khalil.rouhana@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 119 24950 OPERA Operators Training Distributed Real-Time Simulations Keywords: High-Performance Computing and Networking, Simulation Technical -Real Time Simulation, CORBA Uses - Physical and chemical process modeling, training operators, chemical, plastics The global objective of the project is to specify, develop and qualify a cost-effective simulator prototype for training of operators in the field of chemical and plastics processes. From the endusers point of view, major expected impacts of the technology shall be : • the improvements of the learning process efficiency, and demonstration of cuts in training time and costs, • the increase of operators productivity, the improved manufacturing performance, the enhanced safety of operations and • the exhibition of the consecutive return-on-investments The technology providers objectives are mainly : • to address new markets, • to extend current product design based on cost-effective technologies and open architecture toward distributed computing technolgoy to address performance requirements as regards to real-time requirements for training applications, and with new models and possibly new services (like openess to Internet/Intranet technologies and • to supply customers with a complete set of software tools in the field of chemical process and control systems simulations. Contact Point Dr. Pascal TROUVE Thomson Training & Simulation Energy Unite Rue du Général de Gaulle 1 Z.I. Les Beaux Soleils, Osny 95523 Cergy-Pontoise, FRANCE Participants TT&S ICI TCL VIA IITB PROSIM Start Date October 1997 Country F UK NL B D F tel +33.1.34.22.83.22 fax +33.1.34.22.86.91 mail: trouve@tts.thomson.fr http: Role C P P P P P Duration 24 months This page is located at www.cordis.lu/esprit/src/24950.htm It was last updated on 29 October 1998, and is maintained by Agnes.Bradier@cec.be Octobre 1998 High Performance Computing and Networking / 120 24959 CRISP-DM CRoss-Industry Standard Process for Data Mining Keywords: High Performance Computing and Networking, Decision Support Technical - Data Mining Process Uses - Engineering, Banking Industrial Applications - Complex Data Analysis The information society with its all-digital information content, and the advent of HPCN technology to support huge databases, presents users with the problem of interpreting vast amounts of data. Although theoretical work and methodological approaches have been published, data mining (or knowledge discovery in databases) at present is more of an art than a well understood reliable process. There exists no well understood, practical data mining process. This fact particularly hinders data mining projects involving huge databases and is seen as barrier to the profitable widespread deployment of HPCN. The project aims to cater for data mining needs of industrial users of huge data warehouses, by providing an industry-neutral and tool-neutral process model. This project will develop a data mining process which is fast, well understood, reliable, and valid across a wide range of applications. Starting from the embryonic knowledge discovery processes used in industry today and responding directly to user requirements, this project will define and validate a data mining process that is generally applicable in diverse industry sectors. This will make large data mining projects faster, more efficient, more reliable, more manageable, and less costly. A widely adopted process should foster the development of a multitude of data mining tools which support it, thereby significantly contributing to promote a profitable use of HPCN technology. A "special interest group" (SIG) of users and suppliers will be formed to broaden the basis for development and testing without sacrificing the efficiency and effectiveness of a small, tightlyfocused consortium. The SIG is a key feature of this project, helping to ensure relevance and applicability of the results, and facilitating dissemination and exploitation. The process model developed by the project will be exploited by the data warehouse vendor and the data mining tool supplier to enhance their product and service offerings. The user partners will exploit the results of the project internally to improve their business intelligence and decision making. Contact Point Jens Heile Hejlesen NCR Danmark A/S Vibevej 20 DK-2400 Copenhagen Participants NCR Danmark A/S Integral Solutions Ltd Daimler-Benz AG OHRA Start Date July 1997 Octobre 1998 tel: +45 38 15 76 26 fax: +45 38 33 15 22 email: Jens.Heile.Hejlesen@Copenhagen.NCR.com Country DK UK D NL Duration 18 months Role C P P P High Performance Computing and Networking / 121 24960 PARROT Parallel Crew Rostering Keywords: High Performance Computing and Networking, Decision Support, Optimisation Technical - Operations Research, Constraint Programming, Linear Programming Uses - Transport Industrial Applications - Airline Crew Rostering, Preferential Bidding Systems The objective of the PARROT project is to provide efficient means to address the highly complex and costly problem of airline crew scheduling. Given the traditional breakdown of crew scheduling into (I) crew pairing: identifying a sequence of duty periods from (a) home-base to (another) homebase, and (ii) crew rostering: assigning a set of pairings to each crew, usually on a monthly basis, PARROT proposes to concentrate on improvement of crew rostering quality and performance by applying a parallel approach, and by developing on promising results in the combination of Operations Research (OR) techniques and Constraint Programming (CP). Identification of the optimal "mix" of techniques and the appropriate degree of parallelization for the characteristics of different airlines will result in approaches that are adapted to small, medium and large-size companies' situations. The goal of PARROT is to produce the following output: • A parallel solver, proven efficient in identified airline crew rostering contexts • A rostering constraint library, for use in crew rostering for airlines and analogous organisations (trains, ...), and conceived for ease of adaptation to neighbouring contexts • Components combining parallelized specific OR- and CP- based crew rostering algorithms, proven efficient in identified airline crew rostering contexts • Agreed recommendations and experimentation with a trans-nationally applicable regulations definition language. • An evaluated and validated demonstrator customised for use at each of the airline sites. ILOG will add a constraint library dedicated to rostering, together with the parallelised constraint solving engine, to its product line. Lufthansa Systems and Carmen Systems will commercialise dedicated airline crew scheduling products. Olympic Airways plans on deploying project results directly in-house. Both universities intend to carry out research and publication activities involving both OR and CP. Contact Point Mari Georges ILOG 9, rue de Verdun, BP 85 F-94253 Gentilly Cedex Participants ILOG, Intelligence Logicielle Lufthansa Systems Carmen Systems Olympic Airways University of Paderborn University of Athens Start Date To be announced Octobre 1998 tel: +33 1 49083557 fax: +33 1 49083510 email: georges@ilog.fr URL: http://www.ilog.fr Country F Role C D S GR D GR P P P A A Duration 30 months High Performance Computing and Networking / 122 24977 STRETCH STorage and RETrieval by Content of imaged documents Keywords: High Performance Computing and Networking Technical: Multimedia documents, Image Analysis, Pattern Recognition, Archival Systems Uses – Object-Oriented paradigm, Parallel Program Development, Document Management Systems Industrial Applications – Accounts payable, Public documents, Medical Images The aim of the project is to develop a pre-production prototype of a system capable of storing and retrieving imaged multimedia documents in document databases, by applying advanced techniques derived from image analysis and pattern recognition. Such techniques already exist, at different degrees of maturity, but are separately employed in current document processing or image archival systems. The intent of STRETCH is to create a common Archival and Retrieve Shell based on a graph document representation and capable of activating appropriate functions to characterize and subsequently retrieve (multimedia) documents on users’ demand. To make such a system effective, the bottlenecks of document profiling must be avoided; in particular by overcoming the existing limitations of predefined indexing schemes. The document database will include intrinsically multimedia documents with an object-oriented internal representation and related characterization and search methods. The reliance on an iconic interface will allow use of documents by an international community. The consortium intends to exploit the results of the project by enlarging the offers of document management systems, so far limited to indexed image archival and ICR of fixed form types. Besides, the content-based retrieval by images will open new application fields. Contact Point AnnaMaria Colla Elsag Bailey – Finmeccanica SPA Via G.Puccini 2 16154 GENOVA, Italy Participants Elsag Bailey – Finmeccanica SPA CEA/LETI AET EBI DPCI O2 Technology DIBE – University of Genova DSI –University of Florence DFC –University of Florence PSI - University of Rouen Start Date December 1997 Octobre 1998 tel: +39 10 658 2516 fax: +39 10 658 2694 email: annamaria.colla@elsag.it Web: http://www.aetnet.it/stretch/index.htm Country I F I I F F I I I F Role C P A A P P S S S S Duration 30 months High Performance Computing and Networking / 123 24986 DEDEMAS Decentralised Decision Making and Scheduling Keywords: High Performance Computing and Networking Technical : Global Production Processes Optimisation Local Knowledge : Decentralised Control, Communication of Autonomous Manufacturing Entities Industrial Applications : Supply Chain Processes, Multi-Site Production, Interacting Units The motivation of the project is based on the changes in the business environment of manufacturing enterprises. As the pace of change in market conditions, technologies, legislation and raw material availability has accelerated, industrial companies have been forced to improve their business processes and decision making towards agile planning and co-ordination technologies. Currently, there is a strong trend in manufacturing enterprises to move from large, hierarchical organisations to small, decentralised and partly autonomous production units connected with horizontal business processes. • Development of mechanisms and a support system for global production process optimisation based on decentralised operation • Establishment of a framework for combination of local knowledge and decentralised control with communication, co-operation and global goals propagation in order to reach globally optimal production processes. • Establishment of a framework for the communication between autonomous, co-operative manufacturing entities • Improved optimisation when allocating of customer orders and production resources • Improved supervision capability during production • Transparent communication globally and locally for the use of process related knowledge Contact Point Terho Kärkkäinen WM-data Consulting Oy Ilmarisentie 20 FIN-15200 Lahti Finland Participants WM-data Consulting Oy CCP - Centro de CIM do Porto Werner & Pfleiderer Lebensmitteltechnik Ahlstrom Machinery Corporation Institut für Fertigungstechnik und Spanende Werkzeugmaschinen Rautaruukki Oy Metform Technical Research Centre of Finland Start Date First half of 1998 tel: +358.40.5887935 fax: +358.3.8832510 email: tekar@wmdata.com Country SF P D Role C P P SF P D P SF SF P P Duration 30 months This page is located at www.cordis.lu/esprit/src/24986.htm It was last updated on 8 septembre1998, and is maintained by franck.boissiere@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 124 25000 HIPSID High-Performance Simulation for Interactive Design Keywords: High Performance and Networking Technical: Computer Aided Engineering, Parallel Computing, Simulation, Acoustic Design Uses: Mechanical Design, Structural Integrity Industrial Applications: Aerospace, Automotive, Manufacturing Computer simulation is now recognised as an important component of the design process for high technology manufacturing industries. Reducing the time and cost of such simulations is therefore essential to maintain Europe’s competitive advantages in high technology manufacturing industries. The HiPSID project aims to address these needs by creating a new design and simulation environment, combining new, state-of-the-art numerical methods, cost-effective parallel high performance computing (HPC) technology and direct coupling of simulations with CAD. State-of-the-art numerical methods and high-performance computing technology will allow a HiPSID user to simulate an initial design rapidly, and the simulation will be responsive to changes specified subsequently by the user via the CAD interface, for all but the largest problems. The interactive nature of the HiPSID environment will radically alter the role of simulation in the design process, by making it easier for designers to understand and use. The HiPSID environment will be demonstrated using test problems from the automotive and aerospace component industries and the consumer durables industry. The HiPSID environment will be based on the BEASY boundary element analysis package from Computational Mechanics, with CAD interfaces provided by SimTec and Ikerlan. The results of the project will be integrated with existing software products and marketed by these companies following successful demonstration of the prototype system in the HiPSID project. Contact Point Dr Robert Adey Computational Mechanics Ashurst Lodge Ashurst Southampton SO40 7AA Participants Computational Mechanics Fagor/Ikerlan PAC Sener Simtec UPC-CEPBA Start Date December 1997 tel: +44 (0)1703 293223 fax: +44 (0)1703 292853 email: radey@beasy.com http://www.beasy.com Country UK E UK E D E Role C P P P P P Duration 30 months This page is located at www.cordis.lu/esprit/src/25000.htm It was last updated on 29 October 1998, and is maintained by Khalil.Rouhana@cec.be Octobre 1998 High Performance Computing and Networking / 125 Participants Computational Mechanics Fagor/Ikerlan PAC Sener Simtec UPC-CEPBA Start Date December 1997 Octobre 1998 Country UK E UK E D E Role C P P P P P Duration 30 months High Performance Computing and Networking / 126 25009 DOMINOS Domain Decomposition Methods for Integrated Noise and Structural Analysis Keywords: High Performance Computing Technical - Parallel Platforms, Vibro-Acoustic Prediction Uses – Computer-based Numerical Analysis, Finite Element Method Industrial Applications – Automotive Noise & Vibration Development, Compressor Noise The solution of large structural-acoustic coupled problems in a reasonable time requires the availability of advanced and optimized solution strategies adapted to currently available and possible future HPCN platforms, and especially to various kind of parallel architecture. The domain decomposition method has proved over the last few years to an extremely attractive parallelization techniques for large-scale static and dynamic elasticity problems. Similarly, parallelized conjugate gradient techniques have had little applications outside linear statics. Application of these two techniques to structural-acoustics is therefore a true challenge. Vibro-acoustic simulation has evolved over the last few years from a research tool to a widely accepted design tool in areas as diverse as automotive and aerospace engineering, home appliance design, audio equipment engineering and defense applications. Europe clearly leads the way in this field where the project coordinator is recognized as the premier supplier over Europe, America and Asia. Vibro-acoustic simulation being an extremely computationally intensive technology, the need to apply domain decomposition principles is therefore clear. The primary goal of this project is to demonstrate the versatility of domain decomposition methods and to produce significantly faster and powerful vibro-acoustic simulation tools for the benefit of the European industry. Contact Point Pierre Guisset LMS International Virtual Product Refinement Interleuvenlaan 68 B-3001 Leuven Participants LMS Numerical Technologies Hutchinson SA ONERA APEX Technologies ULB, Brussels Daimler-Benz AG ABB Corporate Research Start Date October 1997 tel: +32 16 384 500 fax: +32 16 384 550 email: pierre.guisset@lms.be Country B F F F B D S Role C P P A A P P Duration 30 months This page is located at www.cordis.lu/esprit/src/25009.htm It was last updated on 29 October 1998, and is maintained by Khalil Rouhana@cec.be Octobre 1998 High Performance Computing and Networking / 127 25047 QUASI Quantum Simulation in Industry Keywords: High Performance Computing and Networking, Simulation Technical: Quantum Mechanics, Molecular Mechanics Uses – Modelling, Simulation Industrial Applications – Catalytic Chemistry The aim of the project is to extend and implement state-of-the-art techniques for combined Quantum Mechanics/Molecular Mechanics (QM/MM) simulations on a variety of High Performance Computing (HPC) platforms, and to apply the techniques to industrial catalytic chemistry applications. The QUASI project combines software development work of the academic groups with demonstration and applications work from the modelling teams within three major European chemical industries. Three challenging molecular modelling problems in industrial catalytic chemistry have been chosen as themes for demonstration and application calculations, - Heterogeneous catalysis of nitrous oxide decomposition using metal-substituted zeolites, - Modelling of covalently bonded protease inhibitors, - Catalysis by metal oxides and metal supported oxides. One of the objectives is also to analyse the potential market for the QUASI simulation package within the European process industry to prepare for the commercial exploitation of the package. Contact Point Paul Sherwood Daresbury Laboratory Keckwick Lane WA4 4AD Warrington, United Kingdom Participants CCLRC - Daresbury Laboratory Norsk Hydro BASF University Zurich Royal Institution ICI Start Date January 1998 Octobre 1998 tel: +44 1925 603553 fax: +44 1025 603634 Email: p.sherwood@dl.ac.uk Country UK N D CH UK UK Role C P P P P P Duration 36 months High Performance Computing and Networking / 128 25050 JULIUS Joint Industrial Interface for End-user Simulations Keywords: High-Performance Computing and Networking Technical - Simulation, Mesh-generation, Visualisation, Data Management Uses - Engineering, Design Industrial Applications - Electromagnetic, Aerospace, Automotive The project extends the computing environment and tools developed in the CAESAR ESPRIT project. The backbone of the project is an agreed industrial test case suite of problems of increasing complexity and leads to the final demonstrators of the industrial partners. The demonstrator suite is used to steer and test the developments, which include the HPCN Simulation and Design Environment, Engineering Simulation Tools and the innovative software-based Active Engineering Simulation Advisor Modules and lead to the final industrial demonstrators. The developments are aimed at providing a balanced simulation environment whilst addressing increasing complexity and larger problems, reducing pre and post-processing bottlenecks and enabling substantial increases in design effort / unit time via the provision of an integrated simulation and design environment, called 6S (Sixth Sense), and toolset for HPCN simulation in multiple disciplines. Contact Point Mr. John Murphy tel +44-177-936 6260 British Aerospace, Sowerby Research Centre fax +44-177-936 3733 PO Box 5 email: john.murphy@src.bae.co.uk UK - Filton, Bristol, BS12 7QW http: to be announced Participants British Aerospace DASA Dassault ESI Stehlin-Merazzi Research Genias FhG-IPK NAG Ltd. University of Oxford University of Swansea Swiss Centre for Scientific Comp. Start Date 1st February 1998 Octobre 1998 Country UK D F F CH D D UK UK UK CH Duration 30 months Role C P P P P A A A A A A High Performance Computing and Networking / 129 25058 DECISION HPCN Integrated Optimization Strategies for Increased Engineering Design Complexity Keywords: High-Performance Computing and Networking Technical - Optimisation, Decision Making Uses - Engineering, Design, Multidisciplinary Analysis Industrial Applications - Aeronautics, Materials, Machinery DECISION aims to set up co-operative work on design optimisation problems of increased complexity. The combination of single or hybridised constrained optimisation tools with analysis solvers assembled on a HPCN platform targets innovative multidisciplinary engineering products. DECISION is a decision maker oriented tool for new product design optimisation. It provides an emergent technology for industrial competitiveness by significantly reducing design cycle and cost for new products. Complex design problems coming from three main industrial sectors - Aeronautics, Smart Materials and Forest Machinery - have been chosen to validate the efficiency of the design optimisation system and to provide quality with background of a wide range of optimised results. Contact Point Mr. Jacques Périaux tel +33.1.47 11 37 89 Dassault Aviation fax +33.1.47 11 42 94 78, Quai Marcel Dassault email: dpr.dir@Dassault-aviation.fr 92214 Saint Cloud Web: http://www.cooperation.dassaultaviation.fr/index.htm FRANCE Participants Dassault Aviation Messet Noka Tume NAG Ltd. INRIA University of Jyvaskyla VTT Start Date October 1997 Octobre 1998 Country F SF SF UK F SF SF Duration 30 months Role C A P P A A A High Performance Computing and Networking / 130 25059 CHAMAD COMPLETE HEXAHEDRAL AND ADAPTATIVE MESHING OF ARBITRARY DOMAINS Keywords : Simulation, Pre-processing, Automatic Meshing Technical : Mesh generation, Mesh refinement Uses : Engineering and scientific computing, Industrial Applications : Automotive , Aerospace, Chemical industries Summary Pre-processing phases and more specifically mesh generation is recognised to refrain the use of HPCN analysis, simulation and design in a range of applications in industries such as automotive mechanical engineering, chemical engineering, aeronautical engineering. The objective of this project is to automate mesh generation and mesh adaptation and especially hexahedral unstructured grids. The project is responding to strong requirements already expressed by end-users from automotive, aeronautical and civil engineering fields who have identified the bottlenecks in their design cycles and who want to benefit from the considerable progress made in the computation and visualisation parts of the process. The project is built on existing prototypes for both automatic mesh generation and mesh refinement which are being evaluated by the end-users. The Consortium is composed of two major industrial end-users (SNECMA and BMW), three software vendors, one service provider and one CAD system vendor, from Belgium, France, Germany, UK and Norway. The planned developments will be validated on a large set of cases covering many industrial applications fields in computational electromagnetics, fluid dynamics, thermal analysis and structural analysis. Contact Points Laurent LEQUETTE BERTIN et Cie Systèmes d'Information et Modélisation Avancée 59, Rue Pierre Curie, BP 3 F- 78373 Plaisir Cedex Participant BERTIN BMW NUMECA SAMTECH SNECMA VECTOR FIELDS VIEWTECH Start Date 1 December 1997 Tel: 331.34.81.87.12 Fax: 331.34.81.87.43 Email: lequette@bertin.fr Country F D B B F UK N Role C A P P A P A Duration 24 months This page is located at www.cordis.lu/esprit/src/25059.htm It was last updated on 29 October 1998, and is maintained by Khalil.Rouhana@cec.be Octobre 1998 High Performance Computing and Networking / 131 26255 Gabby Animator How to rise the productivity in the animation market Keywords: High Performance Computing and Networking Technical: 3D animation, Visualisation, Real-time technology Industrial Applications : Animation series The aim of the project is to develope a software package based on applications of real-time technology to 3D animation that will make it possible to produce a new kind of animation series more quickly and at lower cost. The goals will be achieved through the use of real-time animation and visualisation tools that will make it possible to generate complex animations, such as expressing emotions through a characters face or body movements, an visualise and replay these animations as they are being generated. The tools developed will enable animators to achieve enhanced productivity through the application of new technology to 3D animation. The link with traditional 2D production will occur only through "clear line" rendering, giving a final result similar to that produced by classic pen-andpaper techniques. The industry version of Gabby Animator will be used within the companies that participate in its development to cut production costs and times and increasing the quality of the output. A "Home" version – a plug-in for 3D Studio Max – will be exploited through the distribution networks of Kinetix and Autodesk. Contact Point Sebastien Dohogne Neurones Animation 53, Quai de Rome 4000 Liege, Belgium Participants Neurones Animation Luxanima Laboratoire d’Imagerie (LIN/CNBDI) Start Date May 1998 tel: +32 4 254 61 12 fax: +32 4 229 32 33 Email: seb@neurones.be Numerique Country B L F Role C P P Duration 24 months This page is located at www.cordis.lu/esprit/src/26255.htm It was last updated on 1 July 1998, and is maintained by pertti.jauhiainen@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 132 26267 SCHUMANN Supply Chain Uncertainty Management Network Optimisation Keywords: High Performance Computing & Networking, Decision Support Technical: Optimisation through Recursion, High Level Model Generation, Automatic Uncertain Data Scenario Generation, Multi-site Analytic Decision Data Warehouses Uses: Simulate/Optimise Supply Chain for Manufacturing, Assembly, Distribution & Service Industrial Applications: Automotive The project aims at developing, implementing and testing, on a high performance computing platform, a software system to simulate and optimise the Supply Chain for Manufacturing, Assembly, Distribution and Service (MADS) of Bills of Material by using the automotive sector as a pilot area. The target is to develop and implement mathematical algorithmic approaches jointly with suitable analytic decision data warehouses for a multi processor parallel high performance computing environment, so that the solution times for a large size case of the above problem lie in the range of 15-30 minutes for a rough cut and 3.0-3.5 hours for detailed allocation and scheduling/planning. The currently available systems are not capable of solving the problem optimally. They just provide implementable solutions by addressing the problem as a sequence of subproblems, for each part of the supply chain and each major supplier at each time period along the planning horizon. Although the computing effort is affordable with present systems, however they do not guarantee the optimality of the solution. The proposed project addresses improvements in both quality (better allocation, lower costs) and efficiency (shorter response times) in MADS Supply Chain scheduling through parallel computing algorithm implementations. This will have a direct bearing on the industrial partners’competitiveness world-wide as they face the challenges of market globalisation. It is estimated a 15% output increase from the current capacity, a 20% product’s lead time reduction and a 15% inventory cut-off by using the SCHUMANN software result. It turns out in money savings (up to 25 MECU in some levels of the supply chain). Furthermore, the software developments which result will be of benefit to other industrial end-users beyond the SCHUMANN consortium. Contact Point Laureano Escudero UITESA Avda. de Burgos 8B E-28036 Madrid Participants UITESA LPC DAIMLER-BENZ, AG FORD-Espana Brunel University Universidad Politecnica de Valencia Start Date January 1998 Octobre 1998 tel: +34 1 383 31 80 fax: +34 1 383 33 11 email: leb@uitesa.es Country E UK D E UK E Role C P P P A A Duration 30 months High Performance Computing and Networking / 133 26276 SEP-TOOLS AN INTER-OPERABLE SUITE OF EUROPEAN HPCN TOOLS Keywords : Software Development Environments, Parallel processing Technical : Performance prediction, Code parallelisation Uses : Engineering and scientific computing Industrial Applications : Automotive , Aerospace, Weather forecast Summary The SEP-Tools project will develop a suite of HPCN software tools that support parallel code development in Fortran using MPI. The suite is based upon well established European software tools. Tools in the suite will be interoperable and the suite will be deployed on leading-edge HPC systems from the two largest Japanese HPC vendors (Fujitsu and NEC) in order to encourage the international take-up of developments. To further strengthen the exploitation routes for the suite, a major European HPCN software house will co-ordinate the project. The starting point for SEP-Tools are the following core software technologies: − − − DIMEMAS: a novel performance prediction tool for message-passing applications; VAMPIR: a performance visualisation tool for message-passing applications supporting MPI; CAPTools: a semi-automatic code parallelisation tool capable of complex code dependency analysis. It is important to note that whilst each of these tools is individually useful in the porting and migration of existing Fortran codes, it is the ability to fully inter-operate that will provide the application developer with the opportunity to develop new ways of working, further enhancing his productivity. Since all the tools included in the suite have already been developed, the primary focus of the project will be in developing the interfaces necessary between the tools to ensure interoperability. This substantially reduces the risk of the project since effort will be only spent on the later stages of tool development rather than on developing new tools. This will also enable the project to be completed over a relatively short time-scale (18 months). The SEP-Tools consortium promotes a partnership between three of Europe’s leading software tool developers (CEPBA, PALLAS and University of Greenwich) with key software tool exploiters (Fujitsu, NEC and PALLAS) and end-users (LMS, PDC and Danish Meteorological Institute). Furthermore, in order to ensure that the tool suite addresses the requirements of the widest possible spectrum of HPC users and code developers a project “User Forum” will be established. Contact Points Mr. Karl Solchenbach Hermülheimer Staße 10 D-50321 Brühl Germany Tel +49-2232 1896-0 Fax: 49-2232 1896-29 Email: solchenbach@pallas.de Dr. Owen Thomas Smith Group Surrey Research Guildford GU2 5YP UK Tel: +44 1483 442118 fax: +44 1483 442304 Email: OGMThomas@smithgroup.co.uk Participants Country Octobre 1998 Role High Performance Computing and Networking / 134 PALLAS FECIT NEC CEPBA University of Greenwich KTH DMI LMS Start Date D UK D E UK S DK B C P P P P A A A Duration 18 months This page is located at www.cordis.lu/esprit/src/26276 It was last updated on 29 October 1998, and is maintained by Khalil.Rouhana@cec.be Octobre 1998 High Performance Computing and Networking / 135 26285 CAVALCADE COLLABORATIVE VIRTUAL CONSTRUCTION AND DESIGN Keywords: High Performance Computing and Networking Technical: Distributed Virtual Reality, groupware, multi-modal interface, cross-platform Uses: Virtual Mock-up, teleconferencing, PCs, 3D interaction devices Industrial Applications: architecture and construction, car design and manufacturing, satellite and railway modelling FThe CAVALCADE project goal is to develop a computer system for collaborative prototyping of virtual mock-ups. It will support and enhance concurrent engineering practices thanks to a distributed architecture enabling teams based in geographically dispersed locations to collaboratively design, test, validate, and document a shared model. CAVALCADE will provide users with a 3D visual simulation system where multiple input modalities will simultaneously enable real-time multi-user interaction with a prototype by pointing, looking, asking questions, and issuing commands. Object creation, cloning and deletion, collision detection, assembly trajectory planning, direct and inverse kinematics, as well as Newtonian dynamics simulation will be under direct user control, while high level declarative object placement functions will speed up the design process. Support for standard file formats such as Catia will ensure interoperability with CAD systems. Interactive selection of prototype components will trigger multimedia data access (e.g. text, diagrams, web links). The 3D model thus becomes a visual information query front-end, making CAVALCADE a suitable data management tool for project design. It will be modular and cross-platform, enabling users to choose a configuration depending on the budget. CISI coordinates the project and develops CAVALCADE core (multi-modal dialogue management, declarative functions recognition, teleconferencing support, etc.). IRIT develops a software layer to support cooperative work and multisite data distribution. Sense8 Europe extends its cross-platform development toolkits to fit project needs, especially to provide interoperability with CAD systems. CRS4 develops time-critical components in order to support effective 3D multi-user interaction. AIS develops a rear-projection screen display device for 3D interactive imaging. CSTB, SEAT, CNES, and SNCF, as end users, provide the industrial setting in domains such as architecture and construction, car design and manufacturing, satellite and railway design for specifying and testing CAVALCADE on real-world problems. Contact point CISI 13, rue Villet BP 4042 31029 TOULOUSE FRANCE Tel: +33 (0) 561 176 536 Fax: +33 (0) 261 348 415 E-mail: DuchonJ@sudouest.cisi.fr http://www.cisi.fr/cavalcade Participants Country Role CISI AIS Sense8 CRS4 IRIT FFNS CSTB SEAT CNES SNCF F P P P P A A A A A C I CH I F S F E F F Start date Duration January '98 24 This page is located at www.cordis.lu/esprit/src/26285.htm It was last updated on 1 July 1998, and is maintained by Merce.Griera-I-Fisa@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 136 26286 VINE Video On Demand and Interactive Video Editing System Keywords: High Performance Computing and Networking Technical: Real-time Multimedia server, Video encoding, Video editing Uses – PC architectures, Compression techniques (MPEG2) Industrial Applications – Video broadcasting The VINE project aims at building a Real Time Multimedia Server addressing the market of professional Video-On-Demand (VOD), Video Editing and Interactive Television. The VINE Multimedia server will be designed with a scalable architecture allowing to deliver up to 768 video streams in a single PC rack, to keep the cost per stream at a low level and gain market acceptance. The server will exploit MPEG2 technology for image delivery to guarantee the highest quality. It will include facilities to allow its users: • to edit, develop and archive their own videos, • to access in real time to any video or multimedia material stored on the server, • to store with to day's standard disk technology up to 320 hours of MPEG2 videos. Communication will be handled over high bandwidth networks (such as Fast Ethernet, ATM OC3 or ADSL modems) and implement a very high quality Multimedia Intranet system. The VINE server targets professional markets such as: • TV stations to address the needs of the journalists who want immediate access to large image banks, edition facilities and digital quality to better suit the requirement of the public, • Corporate networks to address the needs of training and informing for employees, franchisees and customers about new products and new services, • Small to middle size Video-On-Demand networks that can be installed in hotels or districts. As part of the developments, VINE will also deliver a new highly innovative MPEG2 encoding station implementing a single pass variable bit rate algorithm to transform in real time professional videos (D1) into MPEG2 and to provide a multicasting facility also allowing broadcast of live television to the users. The VINE technology will be validated and demonstrated by: • The journalists of ORF Upper Austria in Linz who will use it to edit, store and retrieve news clips produced everyday for broadcasting, • The visitors of the Ars Electronica Centre, in the "Museum of the Future" in Linz, to give them a free access to large video material databank and to evaluate its use for educational purposes. Direct exploitation of the results will be made by marketing the VINE server on a world-wide market, targeting telecom and cable operators but also professional applications such as those to be developed in the project. Contact Point Michel Loi tel: +33 4 72 72 86 53 Matra Systemes et Information fax: +33 4 72 72 04 67 6, Rue Dewoitine Email: mloi@matra-ms2i.fr 78142 Velizy-Villacoublay Cedex, France http://www.arttic.com/projects/VINE Participants Country Role Matra Systemes et Information F C Optibase Ltd. ISR P Österreichischer Rundfunk (ORF) A P Start Date Duration April 1998 24 months The URL of this document is http://www.cordis.lu/src/26286.htm It was last updated on 1 July 1998, and is maintained by pertti.jauhiainen@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 137 26287 M3D Multisite Cooperative 3D Design for Architecture Keywords : High Performance Computing and Networking, Networked Architecture Design, 3D CAD, Cooperative Work, Virtual Studio, Multimedia Interactive Communication Service. Technical domain : High Performance Networking, Distributed Computing, Multi-Site Application, Persistent Database, real time system, CSCW. Industrial Applications : Architecture design, building construction and maintenance Synopsis The objective of the project is to integrate the CAD technology with the VR and high performance network technology to be a prototype multi-site cooperative 3D system for architecture design. The foreseen prototype system will extend the major functions of the single user CAD architecture design tools to a networked cooperative system at multiple locations. Real time interactive cooperative design with audio/video conferencing will be realized in the system with other off-line functions. The current target geographic area to cover is at the European scale. It can be extended in the future. Contact Point Professor Yuhua Luo Math and Computer Science Department University of Balearic Islands Cra. Valldemossa km 7,5 07071 Palma de Mallorca, Spain Participants Country UIB Spain ADETTI Portugal EDC Netherlands Portugal OA IDOM Spain ARQMAQ Spain Start Date: April 1, 1998 Fax: 34 71 173003 Tel.: 34 71 173208 Email: dmilyu0@clust.uib.es http://www.m3d.org/ Role C P P P P P Duration: 36months The URL of this document is http://www.cordis.lu/src/26287.htm It was last updated on 1 July 1998, and is maintained by Merce.Griera-I-Fisa@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 138 26293 MINEREC Demonstration of real-time Mine Recognition Keywords: Mine Detection, Ground-Penetrating Radar Technical - Digital Signal Processing, High Performance Computing Uses - Demining Industrial Applications - Demining systems, Real-time Applications, Embedded Systems Objectives of the work programme are • To design, develop and test a suite of software comprising known algorithms and data collection procedures for the specific purpose of detecting and recognising anti-personnel mines • To implement the resulting software suites on the EUROPRO architecture and test its functionality • To Design and build a radar array and demonstrate real-time detection and recognition of plastic anti-personnel mines, in laboratory conditions. The work programme will result in the establishment of design parameters of radar/EUROPRO hardware providing design data for operational systems Impact Presently, the realisation of effective ground probing radar systems to detect and recognise plastic anti-personnel mines is frustrated by the incidence of high false alarm rates. Recognition procedures require very intensive data processing, which until the emergence of the EUROPRO architecture has not been available. The realisation of real time recognition procedures and the resulting decrease of false alarms will be a major contribution to the design of operational systems. Contact Point Richard Chignell Emrad Ltd Unit 6 Langham Park/Catteshall Lane UK - GU7 1NG Participants Emrad Ltd Thomson Marconi Sonar Start Date January 1998 Octobre 1998 Country UK F tel: +44.1483.418600 fax: +33 1483.418618 email: 73064.226@compuserve.com Role C P Duration 18 Months High Performance Computing and Networking / 139 26313 CIPRESS Complex Information Patterns Retrieval with a Parallel Distributed Processing Knowledge Engine Search System Keywords: High Performance Computing & Networking, Tools for Information Handling Technical: Knowledge Engine, Free Text Retrieval and Database Engine, Parallelisation, Association Process, Learning Process, Neural Networks Uses: Knowledge Engine Search System Industrial Applications: Medical The objective of the CIPRESS project is to develop a Knowledge Engine with the capability to sort out relevant information from textual and numerical information containers such as databases and the Internet by first discovering facts in documents and reports (the association process) and by keeping track of previous searches and the organisation of hierarchical thesauri (the learning process). This CIPRESS Knowledge Engine will be developed starting from an existing high performance generalised free text retrieval and database engine, to be parallelised by the introduction of a scalable High Performance Architecture to further boost its performance and upon which will be developed: (i) an analyser that will convert into a standard hierarchical format the data collected by the knowledge engine and create associations between them, (ii) a sophisticated neural network platform allowing to continuously improve the use of the Knowledge Engine and refine more accurately the data to be collected by the analyser and its users, and (iii) a user interface, allowing to facilitate and optimise exploitation of the Knowledge Engine features. The first application field to be considered will be the medical sector where numerous databases are available (Medline, CANCERLIT, DrugLine, etc.). This sector has also the advantage of representing a very large market already mature for the provision of information services. The CIPRESS Knowledge Engine applied to the medical sector will be widely exploited through commercial services for doctors, for pharmacists, for scientific purposes and for the industry. The CIPRESS Knowledge Engine will be commercialised by Search&Find, while all partners will exploit the project results by setting-up on-line services in the medical sector. Contact Point BO LINDSTRÖM SEARCH&FIND FINLANDSGATAN 14 S-16474 KISTA Participants SEARCH&FIND ALEPH INFORMATICA Karolinska Institutet - MTC ARAKNE HealthGate Europe Start Date To be announced Octobre 1998 tel: + 46 8 751 84 40 fax: + 46 8 752 05 30 email: BO.LINDTROM@CONTACTOR.SE Country S I S I UK Role C P P P A Duration 24 months High Performance Computing and Networking / 140 26317 EUROSTORE A High Performance Storage Project Keywords: High Performance Computing and Networking Technical – PFS, Parallel File System, HSM, Hierarchical Storage Management, Storage, Data Models, Metadata, Near-line Storage, Storage Robotics Uses – Digital Movies Storage, Large Databases, Data Mining, Medical Imaging Storage The objective of the EUROSTORE Project is to develop and market a High Performance File Store, composed of an Hierarchical Storage Management System (HSM) operating in conjunction with a high performance Parallel Filesystem (PFS). The Project will combine the features of a hierarchical storage manager component with the performance of a parallel filesystem to provide a system capable of meeting the requirements of the most demanding applications in industry, commerce, and science, where the basic need is the management of very large data repositories, with transparent access to both on-line and near-line storage. The Project is motivated by the fact that the few existing HSM and parallel filesystem products do not fully address all the needs of the user community, and in some cases represent a major investment that users are not willing to undertake. The EUROSTORE Project will enable the partners to build high performance storage systems from low cost components. It will lay the foundation for commercial products with broad market appeal. Contact Point Carlo Alberto Marchi QSW Ltd. Via Marcellina, 11 I - 00131 Roma Participants Quadrics Supercomputers World Hellenic Company for Space Applications DESY CERN Star System Roma Hellenic National Meteo Center Athens Medical Center Start Date March 1998 Octobre 1998 tel: +39-6-412386.16 fax: +39-6-4191694 email: ca.marchi@roma.quadrics.com http: to be announced Country UK GR DE CH I GR GR Role C P P P A A A Duration 24 months High Performance Computing and Networking / 141 26329 MISSIMU Minimally Invasive Surgery Simulator Keywords: High-Performance Computing and Networking, Technical : Finite Element Models, Parallel Computing, 3D Graphics, Mechatronic Uses: Medical Simulation Industrial Applications : Endo-surgery Training, Operation Preparation and Rehearsal, Assessment of new surgical tools The MISSIMU RTD project aims at specifying, developing and qualifying a simulator suitable for the training of surgeons in endo-surgery, or "Minimally invasive" techniques. This simulator is intended for students or inexperienced surgeons for acquisition of basic and advanced skills, experienced surgeons for the rehearsal of complex operations and tool makers for the assessment of new surgical tools. Salient features of the simulator are : -modelling of organs in real time taking into account trainee actions, as well as organs mechanical characteristics -generation and display of realistic computer generated images comparable within reasonable tolerances to those generated by endoscopic camera, -restitution of tactile feel trough appropriate force feedback system, -capacity of providing remote training through a WAN. MISSIMU is based on the expertise in modelling, simulation, 3D graphics, mechatronic technologies and surgical tooling of the technical partners as well as on the deep knowledge of operational requirements of end-users which are hospitals and medical universities. MISSIMU has the potential to demonstrate the applicability of simulation techniques to surgeon training contributing thereby to the spread of minimally invasive practices and paving the way to the development of proficiency standards. Contact Point : Dr. Klaus-Martin Irion Karl Storz GmbH & Co Mittelstrasse 8 D - 78532 Tuttlingen Tel: +49 7461 708 219 Fax: +49 7461 708 304 e-mail: karlstorz-webmaster@karlstorz.de Participants Karl Storz GmbH Thomson Training & Simulation University of Bristol University of Dundee Assistance Publique et Hôpitaux de Paris Ospedale Nuovo Start Date February 1998 Octobre 1998 Country D F UK UK F I Role C P P A A A Duration 36 months High Performance Computing and Networking / 142 26331 DREAM Data fusion as a Remedy Against Mines Keywords: Mine Detection, Data Fusion, Human Factors Technical - Data Fusion, Human Factors, High Power Computing Architectures Uses - Demining Industrial Applications - Demining systems, Real-time Applications, Embedded Systems In mine detection research, various sensor technologies are currently being investigated, and, although each solution offers interesting possibilities, it is already clear that no single technology can offer both high detection performance and the level of efficiency required. By associating several types of sensors and using data fusion techniques to cross-cue sensor data, both performance and efficiency can be improved. Such a multisensor system, a priori mounted on a vehicle, will reduce the risk and the cost of demining operations. The objectives of this project are to anticipate the development of such new concept of humanitarian demining system. The study « DREAM » will technically define a multisensor fusion system based on the results achieved within the feasibility study « CIMIC » awarded under Tender No 96/C 221/10 to this consortium. Within the study all components of the multisensor system, i.e. sensors, dedicated hardware and software for the sensors, the fusion hardware and software, the human factors implementation and the management of external data, will be examined for optimisation. The keys Issues will be the Data Fusion, the analysis and implementation of Human Factors and the breakdown of the overall processing system. The partners will address each necessary task to optimise the multisensor mine detection system : Data Fusion / Human Factors / Data Processing / GPR / IR / Metal Detector. The results of the project will consist of a set of technical documents and demonstrations of rapidprototyping implementation of the key algorithms and of the Man Machine Interface. The results will be directly used in priority in the future overall programme driven by EC Contact Point François NIVELLE Thomson-CSF / TME / MDM 23-27 rue Pierre Valette F-92245 Malakoff Cedex Participants TME SA TCO TCAR Daimler-Benz Research Ebinger GmbH TME Ltd ERA Ltd USFA NPA Start Date January 1998 Octobre 1998 tel: +33 149 653 161 fax: +33 149 653 636 email: nivelle@tme.thomson.fr Country F F F D D UK UK NL N Role C P P P P P P P A Duration 12 Months High Performance Computing and Networking / 143 26337 GEODE Ground Explosive Ordnance Detection System Keywords: Mine Detection, Data Fusion Technical - Data Fusion, High Performance Computing Uses - Demining Industrial Applications - Demining systems, Real-time Applications, Embedded Systems The objective of the GEODE project (Ground Explosive Ordnance DEtection system) is to demonstrate advanced architecture and fusion software for multi-sensor detection, localisation and classification of Anti-Personnel Landmines (APL) ; by combining complementary sensors and using innovative processing techniques, GEODE will demonstrate a higher detection probability than currently achieved by existing systems, a lower false alarm rate and a capability to classify the various detected objects. It is widely admitted that more than 100 million mines have been laid all over the world and that, at the current rate of manual demining operations, it would take hundreds of years to clear the earth; therefore, the requirement for an efficient way to detect, localise and classify mines which will speed up demining while making it safer is obvious. However, the severe operational conditions (variety of terrains, vegetation, APL nature and mining patterns, environment) and the requirement to maintain a high level of performance independently from the observation conditions leads necessarily to a high performance system. Contact Point Gilles Guillemard Dassault Electronique 55 quai Marcel Dassault F-92214 Saint-Cloud tel: +33 1.3481.3293 fax: +33 1. 3481.3104 email: gilles.guillemard@dassault-elec.fr Participants Dassault Electronique SA Emrad Ltd ELTA Electronics Industries Marconi SpA Institut Dr. Foerster TNO-FEL Country F UK ISR I D NL Start Date January 1998 Duration 15 Months Octobre 1998 Role C A A P P P High Performance Computing and Networking / 144 26342 NOVICE Network Oriented Visualization in a Clinical Environment. Keywords: High-Performance Computing and Networking Technical -3D Visualisation, Imaging, Internet, Java/ActiveX, Parallel computing Uses - Web-based Visual Medical Application Industrial Applications - Computer Aided Medical Diagnostics (CAMP) The project will provide extendible Web-based visualisation tools for medical applications in a high performance-computing environment, transparently for the end users. It will improve existing imaging techniques to transport, store, visualise and extract information from images (e.g. parametric images, statistical information). Emphasis is on industry standard networked technologies (Java, ActiveX and VRML), enabling remote consultation with experts, and easy reexamination by physicians. A primary charter of NOVICE is to help the European hospitals to advance the state-of-the-art in medical care. Deploying Computer Aided Medical Diagnostics (CAMD) tools outside the research facilities is a difficult challenge facing all research hospitals. NOVICE will provide a Web-based Demonstrator and Service Center with parallel visual computing, accessible to all European hospitals, where modern CAMD tools can be easily reviewed by the end users. NOVICE brings together the expertise needed to develop such a European Demonstrator. The NOVICE project will deliver a number of industrially relevant and exploitable results, which will have a significant impact on patient care and health service costs across Europe: • Network oriented visualization application components for medical applications • A NOVICE Demonstrator and Service Center at MVC • A parallel library of 3D image processing functions • A teleradiology tool accessed via the Web • An enhanced PACS systems capable of storing full multimedia patient records • Contact Point: Dr Mikael Jern tel: +45 45999599 Advanced Visual Systems fax: +45 45999500 15, Blokken e-mail: mikael@avs.dk DK - 3460 Birkeröd http://www.avs.com Participants Advanced Visual Systems Silicon Graphics Biomedical Sistemas Expertos Central Manchester Healthcare Trust Manchester Visualization Centre Foster Findlay Rasna Imaging Systems Dep of Oncology Univ Pisa Clinica Femenia Hospital San Juan de Dios Start Date January 1998 Octobre 1998 Country Denmark Israel Spain UK Role C P P P UK UK Italy Spain Spain Spain Duration 36 months P P A A A A High Performance Computing and Networking / 145 26347 EDISON European Distributed Interactive Simulation Over Network Keywords : High Performance Computing and Networking, Technical : Distributed Simulation, Collaborative Working, DIS, CSCW, HLA, ATM Uses : Simulation, Training, System Design, System Validation, Engineering Analysis, Crisis Management Industrial Applications : Space, Aeronautics, Automotive, Telemedecine, Telescience, Petroleum Platform Summary: Aim of the EDISON project is to specify, develop, experiment and exploit a generic and integrated architecture for Distributed Interactive Simulation Facilities (DISF), in order to support interaction and cooperative working between geographically distributed facilities for simulation modelling, processing and post-processing. The EDISON infrastructure is a modular package including a simulation framework providing the applications with common simulation services (scheduling, real-time kernel, etc...), a middleware containing the vital services necessary to distributed simulations (time management, intelligent distribution mechanisms, extrapolation and prediction mechanisms to hide network latencies and jitter, etc...), a communication framework, a supervisor as well as groupware functions. This infrastructure is qualified through three pilot applications, each of them focusing on a different utilization paradigm and a different phase in a system life-cycle : the « HardWare-In-the-Loop » pilot application consisting in a system validation of the ATV-ISS space rendez-vous (the Automated Transfer Vehicle is a spacecraft which docks automatically to the International Space Station, in order to refuel and resupply it), the « Numerical Models » pilot application dealing with the engineering analysis of vibroacoustics effects in manned space systems and car design, the « Man-In-the-Loop » pilot application based on a mission rehearsal exercise involving the space European Robotic Arm (ERA) and ATV teleoperations by ISS cosmonauts or ground operators. This DISF approach allows to solve simulation problems only feasible in the past either within long time frames or at the cost of expensive computing support facilities. The use of geographically Distributed and Interactive Simulation Facilities is therefore of paramount interest to reduce facilities overall costs (non-duplication of components, teams knowledge and skills), to allow the interactive use of remote simulation resources, making the work more flexible and efficient, to shorten the development cycle of new systems by anticipated detection of design, integration or operational problems, to minimize travels and long collocation of experts, Contact Point : Name : DUBUC tel: +33 1 34 92 25 98 Company : AEROSPATIALE fax: +33 1 34 92 11 74 Address : 66 Route de Verneuil - BP 3002 email: > francois.dubuc@espace.aerospatiale.fr F-78133 LES MUREAUX Cédex FRANCE http://cec.to.alespazio.it/edison.html Participants Country Role AEROSPATIALE F C ALENIA AEROSPAZIO I P ITALDESIGN I P D3 G P RUS G P FOKKER SPACE NL P SILICON WORLDS F P LIP6 F P DRL G P Start Date: 1/3/98 Duration: 30 months This page is located at www.cordis.lu/esprit/src/26347.htm It was last updated on 1 July 1998, and is maintained by Merce.Griera-I-Fisa@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 146 26356 POEM Parallel Processing of Voluminous EDIFACT Documents Keywords: High Performance Computing & Networking, Information Management Technical: Message and Transaction Processing, Parallel Computing, Performance Testing Uses: EDIFACT Message Processing Industrial Applications: Banking, Retail, Manufacturing The prime objective of the POEM project is to develop a high performance EDIFACT message processor using open scaleable HPCN systems with generic support of different message types. For the identification of parallelisation potentials in processing of EDIFACT messages and the development of an efficient scheduling algorithm, the project will exploit the semantics of the document processing steps as well as the structure of the specific EDIFACT message types. The process architecture of the document processor will be adaptable to any kind of parallel platform or workstation clusters. A definitive aim of the project is to meet the requirements posed by several European banks (from about a 100,000 up to more than 1 million message transactions per hour). The technologies developed within POEM will be relevant not only for financial but also for other domains like retail or manufacturing in the near future as well. With increasing volumes in individual branches plus the increasing size of individual messages by using the respective enhanced international EDIFACT message types these branches will soon be exposed to performance limits. Existing industry specific solutions will have to be replaced by a more general and sophisticated approach like the one being suggested in this project. The two banks involved in this project - ABN Amro and WestLB - will play the key role in the initial exploitation of the project results in the banking sector. The explicit intention of ABN Amro and WestLB is not only to cover today’s market demands but to meet tomorrow’s customer requirements for EDI solutions well in advance to its competitors. This can only be achieved using a high performance solution. The initial circle for addressing a wide range of customers will be Assem Audi´s EDI Users Group, a group of major banks including WestLB using Assem Audi EDI solutions. In parallel, Assem Audi will enhance it’s international market coverage by winning marketing partners and reference customers all over Europe. Contact Point Thorsten Reinheimer ASSEM AUDI Feldstraße 8 D-53340 Meckenheim Participants ASSEM AUDI ABN Amro Westdeutsche Landesbank Girozentrale GMD University of SouthamptonPAC Start Date To be announced Octobre 1998 tel: +49 (0) 22 25-881-0 fax: +49 (0) 22 25-881-201 email: tre@assem-audi.com.de Country D NL D Role C P P D UK A A Duration 30 months High Performance Computing and Networking / 147 26386 DISTAL Distributed Software Access for Large-Scale Engineering Applications Keywords: High Performance Computing and Networking Technical - High Performance Networking, Metacomputing Uses - Computer Aided Design, Distributed Engineering Industrial Applications - Car Industry, Aerospace Industry The increasing user-friendliness of engineering analysis codes is enabling engineers to consider running large-scale automated applications with these codes such as parametric design analyses and design optimisation. To provide the huge computing resources needed for these "meta-aplications", it is possible to distribute the work over a large number of loosely-coupled systems, which may even be geographically distributed. Unfortunately, conventional commercial software licensing methods require the underlying analysis codes to be installed on all the systems which will be used. This raises the cost of distributed metaapplications, and constitutes a barrier against "total resource utilisation" for large-scale applications. The objectives of the proposed DISTAL project are to address this by: • Establishing a business model for "software on demand", which is capable of supporting large-scale meta-applications in computer-aided engineering in; an economic way; • Implementing a prototype agent-based system to support this model; • Experimenting with and validation the model for a renage of possible applications which are relevant to European engineering-based businesses. By demonstrating that these scenarios are viable, DISTAL will make possible a new, on-line market in commercial engineering application software, resources and services. All partners will exploit this market as users, application or computing resource suppliers, or as a supplier of secure systems technology to participants in the market. Contact Point No name yet IKOSS Haagse Schouwweg 8g NL-2332 KG Leiden Participants IKOSS CASA Mannesmann Sachs PAC Baltimore Technologies VCIL MSC GmbH MARC Software Deutschland GmbH Start Date Don't know Octobre 1998 tel: +31 71 579 55 00 fax: +31 71 572 12 77 email: http Country NL E D UK IRL IRL D Role C P P P P P P D P Duration don't know High Performance Computing and Networking / 148 26387 DYNALOG DYNAmic LOGistics for distribution networks Keywords: High Performance Computing and Networking Technical: Constraint Logic Programmming, Parallelisation Uses – Forecasting, Decision making, Planning Industrial Applications – Logistics, Scheduling, Distribution Requirements planning The DYNALOG project aims to develop a dynamic, integrated system to solve a wide range of complex distribution and logistics problems. These problems include the routing of multiple delivery vehicles from multiple depots to locations that have delivery time windows that must be adhered to. The effective planning of logistics (for items such as food delivery) results in cost savings, increased competitive edge and greater customer satisfaction. The DYNALOG system’s functionality and performance can also be used at other tactical/strategic levels, through simulation and "what-if" analysis. The DYNALOG system will offer the following features: • active decision support, via the use of HPCN, to react to real-world events; • the use of two approaches (deterministic and stochastic) to provide reliable solutions to problems where some factors may be uncertain or probabilistic; • on-line demand forecasting from historical data, for use in advanced planning and in cases where sales are not known when the produce leaves the warehouse (this is common when delivering ultra-fresh produce to smaller outlets); • integration into the company’s infrastructure and customisation of the software and hardware configuration to meet end-users’ requirements for functionality and performance. The project results available for exploitation will be the deterministic-needs router, the stochastic router, and the demand forecasting tool. These tools will be of use in companies that have to deal with complicated daily transportation problems. Contact Point Lina Koufokotsiou Macedonian Dairy Industry (Mevgal) 57 100 Thessaloniki, Greece Participants Macedonian Dairy Industry (Mevgal) Cosytec Fiege Holycon National Technical University of Atheens (NTUA) Ouroumoff Informatique Synform Parallel Applications Centre (PAC) Start Date May 1998 tel: +30 391 59 107 fax: +30 391 52 901 Email: anprog@mevgal.gr Web: Country Role GR C F P D P I P GR P F UK P P Duration 30 months The URL of this document is http://www.cordis.lu/src/26387.htm It was last updated on 1 July 1998, and is maintained by pertti.jauhiainen@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 149 26700 ELTRAMOS Electronic market and Trading of Modelling Services and assets for engineering SMEs and institutes Keywords: Electronic Commerce Technical - Decision Support, electronic contracts and tendering, transaction management, distant computer simulation Users - Business re-engineering for SMEs Industrial Applications – Water Industry, Education ELTRAMOS addresses the electronic trading of software and expertise by suppliers to SMEs for simulation modelling of water resources and more generally for hydroinformatics applications. The Project aims at development and validation through demonstrators of a new work environment for small engineering companies. In addition the ELTRAMOS platform will be used to show the impact of electronic trading on the associated market and to lay down the basis for future commercial exploitation. The impact will be auto-evaluated within the Project framework by the participating SMEs in terms of their economical performance. The cultural impact on SMEs and educational aspects related to the use of ELTRAMOS will also be assessed by the academic organisations. It is expected that through a regular basis access to a specific software products running on a remote host processor and to associated expertise would : n give SME users commercial access to a wide range of appropriate tools and expertise without the need to purchase those tools; n enable users to get more immediate access to support on-line; and reduce operating costs for users; n make the software and expertise from suppliers available commercially to a much greater and international market; and reduce distribution costs for the suppliers through reduction of intermediate layer; n stress further competition among providers leading to better quality products. Considerable attention will be given to a credible route to full commercial exploitation of the Project results. Contact Point Marc Erlich tel: +33.4.76334208 Laboratoire d’Hydraulique de France – LHF fax: +33.4.76334239 6, rue de Lorraine email: Marc.Erlich@lhf.fr F-38130 ECHIROLLES Participants Country Laboratoire d’Hydraulique de France – LHF F Cap Gemini France F International Institute for Infrastructure, Hydraulic NL and Environmental Engineering – IHE LM Chapeaux BE HYDRATEC F Studio Galli I Danish Hydraulic Institute (DHI) DK Wallingford Software GB Brandenburg University of Technology at Cottbus D Start Date Duration Second Quarter of 1998 30 months Octobre 1998 Role C P P P A A A A A High Performance Computing and Networking / 150 26749 MIMIC Mining The Internet For Marketing Intelligence Keywords: Marketing Intelligence, Information Discovery Technical - Data Mining, Knowledge Discovery, Internet, Electronic Commerce Uses - Customer Segmentation, Behaviour Analysis, Customised Web Page Content Industrial Applications - Electronic Commerce, Virtual Shopping Malls, Marketing The main objective of the project is to give on-line shopping malls the ability to market their products and services effectively by providing information tailored to individual customers. Such customised and directed marketing communication relies on accurate information about the customer and MIMIC will provide this by developing specialised Internet-capable data mining algorithms and tools. These will allow customer behaviour to be monitored and interpreted, and intelligent decisions about web-page content to be made. A major area of work in MIMIC will be the customisation of existing data mining algorithms and the development of new, Internet-specific techniques for extracting information from Internet sources. The MIMIC system will be based on the Clementine data mining tool but will be modified to include these techniques as well as Internet-specific extensions for data capture, presentation, visualisation and deployment of data mining results. Trials of the system will be carried out in different on-line shopping environments, each offering different product lines and services. Contact Point David Pugh Integral Solutions Ltd. Berk House, Basing View Basingstoke, Hants. RG21 4RG, UK Tel: +44 1256 355899 Fax: +44 1256 363467 email: davidp@isl.co.uk http://www.isl.co.uk/ Participants Integral Solutions Ltd. At Internet Adnet University of Ulster Role C P IRL P Start Date April 1998 Country UK F UK A Duration 24 months This page is located at www.cordis.lu/esprit/src/24950.htm It was last updated on 29 October 1998, and is maintained by Agnes.Bradier@cec.be Octobre 1998 High Performance Computing and Networking / 151 27021 COSMOS Construction Site Mobile Operations Support Keywords : Mobility, Satellite, Wireless LANs, Workflow, CSCW, Mobile IP Summary A workflow application will be developed for managing a construction site’s business processes and will be integrated in order to provide a complete groupware environment. A suitable communication infrastructure will be proposed comprising wireless LANs within the construction sites, and satellite links for the interconnection of the company’s headquarters with the various construction sites. Objectives To build a construction site mobile operations support system, based on an appropriate telecommunication infrastructure for communication links both between the company’s headquarters and construction sites and within construction sites. To develop a workflow application for managing a construction site’s business processes both between the site and company’s headquarters as also within the site itself. To demonstrate and assess the applicability and acceptability of the proposed system, through experiments on a COSMOS-based infrastructure interconnecting the headquarters and construction sites of the users of the partnership. Work Approach The project aims at the development of a workflow application for managing a construction site’s business processes which will be integrated with other advanced groupware components (such as document management, white board and video conference tools), in order to provide a complete groupware environment. A suitable communication infrastructure is proposed comprising wireless LANs within the construction sites, and satellite links for the interconnection of the company’s headquarters with the various construction sites. The satellite links will represent a low-cost, permanent communication infrastructure, that will alleviate the problem of the relocation of construction sites and the lack of local access to existing terrestrial wired infrastructure. At areas where the satellite coverage is inefficient wireless terrestrial links could be used. In addition COSMOS will provide an appropriate middleware platform, to integrate the wireless technologies and the satellite infrastructure, making their use transparent to the IP based applications. Impact of Expected Results The main results of the project are: a workflow application which: a)models the company’s business processes providing an efficient construction site’s operations management, and b) integrates a workflow platform with other advanced groupware components resulting in an advanced groupware environment; a mobile middleware platform that will integrate wireless LAN and satellite infrastructures used in the COSMOS system, making them transparent to the TCP/IP based application software. Notwithstanding their capability of offering generic software solutions, which are capable to be exploited in the global IT market, the results of the COSMOS project are targeted to the specific construction sector. Both large companies and SMEs are intersted in the resulting technology. The industrial needs of efficiently manage planning and control operations in the construction sites by providing project managers with easy access to the company Intranet through mobile appliances will be a crucial factor for the successful exploitation of the COSMOS results. Octobre 1998 High Performance Computing and Networking / 152 Exploitation of Results The structuree of the COSMOS consortium, and the specific partnership it builds present a blending of technical expertise and capability for exploitation which have been selected to ensure the most successful achievements. The partners will make different contributions to the exploitation of COSMOS results, and they will have different benefits and sources of revenue: significant industrial construction users will improve their productivity, increase the efficiency in managing complex construction projects operated in remote sites, and reduce their costs; large industrial providers of wireess network technology; industrial and SME system integrators and application developers, a satellite service provider will benefit by commercialising the COSMOS products (platform and application) and services. research partners with expertise in mobile software platforms and applications will make the COSMOS results known in the scientific community; they will benefit from COSMOS by acquiring knowledge about user requirements and application fields of their methodology. Contact Point Mr. Valentino Cavalli Tel +39 51 30 66 44 Omega Generation S.R.L. Fax +39 51 39 03 10 Via Murri, 39 E-mail: cavalli@omega.it I-40137 Bologna Participants Omega Generation INTRACOM Lucent Technologies Philips Research Laboratories MEDIAGATE GMD-IPSI University of Paisley Coopsette Aktor Start Date 1 April 1998 Country I GR UK UK D D UK I GR Duration 30 months Role C P P P P A A A A This page is located at www.cordis.lu/esprit/src/27021.htm It was last updated on 1 July 1998, and is maintained by Merce.Griera-I-Fisa@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 153 27141 EuroTools Working Group for HPCN Tools Promotion Keywords: High Performance Computing and Networking Technical - Parallelization, High Performance Fortran, PVM, MPI, Object oriented technologies Users - Programming Tools, Programming Environments Industrial Applications -Simulation, Scientific computing, The goal of the EuroTools working group is to increase the use of European HPCN Software both inside and outside Europe. This Working Group aims at encouraging the establishment of collaborations between different partners which are involved in the design, the development, the commercialisation, and the use of European software tools for High Performance Computing and Networking. Concrete results foreseen are the following : • increase the visibility of European Industry and Research in the area of Software Tools for High Performance Computing and Networking, • get end-users informed about the existence and functionality’s of European Tools for HPCN, as well as the potential gain of productivity they can obtain using these tools, • enforce end-users driven development and adaptation of tools. Eurotools will co-operate with existing European projects in the field of HPCN such as the HPCN TTN Network, the HPCnet Network of Excellence and other existing organisations and established users groups. End-users are actively involved in the Working Group through participation to an advisory board set up at the beginning of the project. Contact Point Jean-Louis Pazat IRISA/INRIA Campus de Beaulieu F-35042 RENNES, France Participants INRIA ACE CNUCE-CNR ETH Zurich FZJ GENIAS GMD LUND University NASL ONERA PALLAS Queen’s University Belfast SIMULOG TU Dresden TU Munich VCPC Start Date First half of 1998 Octobre 1998 tel: +33.2.99847214 fax: +33.2.99847171 email: pazat@irisa.fr http://www.irisa.fr/EuroTools/ Country F NL I CH D D D S UK F D UK F D D A Role C M M M M M M M M M M M M M M M Duration 24 months High Performance Computing and Networking / 154 27741 EuroTools Working Group for HPCN Tools Promotion Keywords: High Performance Computing and Networking Technical - Parallelization, High Performance Fortran, PVM, MPI, Object oriented technologies Users - Programming Tools, Programming Environments Industrial Applications -Simulation, Scientific computing, The goal of the EuroTools working group is to increase the use of European HPCN Software both inside and outside Europe. This Working Group aims at encouraging the establishment of collaborations between different partners which are involved in the design, the development, the commercialisation, and the use of European software tools for High Performance Computing and Networking. The main objective of the EuroTools working group is to help end-users and tools developers to communicate and exchange ideas ; we do not to take the leadership in defining and developing tools nor to provide an infrastructure for collaboration. This Working Group will co-operate with existing European projects or organisations in the field of High Performance (HPCN TTN Network, HPCnet Network of Excellence and other existing organisations and established users groups). There are no end-users partners directly involved in this action, therefore we will ask end-users to actively participate to the Working Group through participation to an advisory board that will be set up at the beginning of the project. Concrete results foreseen are the following: • increase the visibility of European Industry and Research in the area of Software Tools for High Performance Computing and Networking, • get end-users informed about the existence and functionality’s of European Tools for HPCN, as well as the potential gain of productivity they can obtain using these tools, • enforce end-users driven development and adaptation of tools. Contact Point Jean-Louis Pazat tel: +33.2.99847214 IRISA /INRIA fax: +33.2.99847171 Campus de Beaulieu email: pazat@irisa.fr F-35042 RENNES, France http://www.irisa.fr/EuroTools/ Participants Country Role INRIA F C ACE NL P CNUCE-CNR I P ETH Zurich CH P FZJ D P GENIAS D P GMD D P LUND University S P NASL UK P ONERA F P PALLAS D P Queen’s University Belfast UK P SIMULOG F P TU Dresden D P TU Munich D P VCPC A P Start Date Duration First half of 1998 24 months This page is located at www.cordis.lu/esprit/src/27141.htm It was last updated on 29 October 1998, and is maintained by Khalil.Rouhana@cec.be Octobre 1998 High Performance Computing and Networking / 155 28144 DECAST Enhanced Design Environment for Industrial Casting Processes on Parallel Computing Platforms Keywords: High-Performance Computing and Networking, Casting Simulations Technical - Parallel Processing, PC networks, Unix networks, Distributed Memory, Numerical Simulations Uses - Foundry and Casting Design and Simulation Industrial Applications - Foundry, Casting, Mould filling, Thermal and Thermal-Mechanical Solidification Objective The project will focus on the extension and integration into a heterogeneous parallel computing environment of existing finite element based simulation software for CFD (Computational Fluid Dynamic) and CTM (Coupled Thermo-Mechanical) problems to be used as efficient tools for solving relevant problems of the European casting industry. These modules will be placed inside the user-friendly DECAST which will allow the geometry and mesh to be defined (taking care of the different requirements of fluid flow, thermal and mechanical analyses) and which will allow easy visualisation of the results obtained. Background / Purpose There is a growing need for the 4.200 foundries in Europe (3.500 in the EU), many of whom are SME´s, to apply better numerical simulation to the complex multi-physics problems arising during the casting process when molten metal is poured into complex moulds and allowed to solidify. Almost all available software concentrates on modelling the flow and the cooling of the molten metal. This gives insight into the casting process but does not take the mechanical behaviour such as the movement of the part with respect to the mould into account, a process that can radically change the cooling characteristics of the casting. By following the mould filling with a coupled thermal-mechanical simulation, not only allows the whole casting process to be modelled more accurately but allows the deformation during solidification, the final shape and the final residual stress distribution to be determined, factors which are essential in evaluating the behaviour and reliability of the component in service. The combination of the improved simulation tools, sophisticated numerical methods and HPCN technology will produce a casting design environment that will compete strongly with existing commercial software that only carry out part of this job. Emphasis and consideration will be placed on the requirements of SMEs who form a major part of the potential market: PC networks will be targeted and a pricing policy set accordingly. By the end of the project a near commercial version of DECAST will be available allowing subsequent sales, marketing and support of the software by Quantech, the industrial IT partner within the project consortium. Contact Point Gino Duffett Quantech ATZ S.A. C/ Jordi Girona Salgado s/n E-08034 Barcelona Spain Octobre 1998 tel: +34 93.204.70.83 fax: +34 93.204.72.56 email: info@quantech.es huantech.es High Performance Computing and Networking / 156 Participants Quantech ATZ CIMNE FERESPE F. Miguel Ros IPPTPAN RUS TEKSID Start Date October 1998 Country E E P E PL D I Role C P A A A P A Duration 30 months This page is located at www.cordis.lu/esprit/src/28144 It was last updated on 29 October 1998, and is maintained by Merce.Griera-I-Fisa@cec.be Octobre 1998 High Performance Computing and Networking / 157 28159 ASRA-HPC Advanced Structural Reliability Analysis on High Performance Computers Keywords : High Performance Computing and Networking optimization, reliability, Technical : response surface, FORM, SORM Uses : optimum and reliable design Industrial Applications : nuclear safety, aerospace engineering, shipbuilding The ASRA-HPC project will result in an integrated engineering software system for the reliability oriented optimum design of mechanical components. The optimum can be found either with respect to weight (cost) under reliability constraints or with respect to reliability under weight (cost) constraints for safety critical structures. The aim is a one-level optimization that is achieved by introducing reliability as the objective or as an additional constraint. On the basis of existing advanced finite element and reliability analysis methods the project will implement two complementary approaches to the optimization problem: general response surface methodology for the reliability/finite element interface and a direct method based on sensitivities for general nonlinear surfaces (FORM/SORM). The objective is to create a user-friendly, HPCN based software system exploiting the computational technology by demonstrating the methods on a variety of parallel platforms including clusters of workstations under Windows-NT and Unix with realistically large, industrial relevant models. The project is based on two well established European commercial codes for finite element analysis (PERMAS) and reliability analysis (COMREL/SYSREL). The industrial partners in the project expect to be able to explore new design strategies by applying nonlinear static analysis and dynamics and at the same time to be able to optimize reliability, i.e. maximize reliability under cost constraints. Selected case studies within nuclear safety, aerospace engineering and shipbuilding will be validated by the industrial partners. The most important aspect for all partners is a further extension of numerical optimization methods in an early design stage, producing lightweight and reliable products. Contact Point INTES GmbH Tel : +49-711-784 990 Rolf Fischer Fax : +49-711-784 99-10 Schulze-Delitzsch-Str. 16 Email: fischer@intes.de D-70565 Stuttgart http : www.intes.de Participants Country Role INTES GmbH Germany M Universitad Politecnica de Catalunya (UPC) Spain P RCP Consult GmbH Germany P Technische Universität München Germany A Institute of Fundamental Technological Poland P Research (IFTR) Alcatel Space Division France A Dynalis France A Forschungszentrum Jülich GmbH (FZJ) Germany A Start Date Duration 1 October 1998 36 months This page is located at www.cordis.lu/esprit/src/28159.htm It was last updated on 29 October 1998, and is maintained by Agnes.Bradier@cec.be Octobre 1998 High Performance Computing and Networking / 158 28168 ALICE Quantitative Flow FIELD Visualisation - QF-View Keywords : High Performance Computing and Networking Technical: CFD, EFD, visualisation, real-time Uses : Validation of experiment Industrial Applications : Combustion chamber design, biological applications The ALICE project aims to develop a Quantitative Flow field Visualisation (QFView) system. QFView is a graphical distributed software environment that integrates experimental and numerical simulation systems. With the advent of powerful and cheaper digital image processing techniques and HPCN technology, it is now possible to extract quantitative information from flow visualization (complex) experiments by means of digital video recording along with laser light sheet, and to perform realtime processing on such data. The combination of flow visualization and digital image processing will lead to (i) multiple measurement capabilities and (ii) cost-effective equipment for Experimental Fluid Dynamics (EFD). The QFView database will reduce the global costs and time required to process and validates measurement data against other experiments or Computational Fluid Dynamics (CFD) results. Contact Point Dean Vucinic Vrije Universiteit Brussel Pleinlaan 2 B - 1050 Brussels Tel : 32-2-629 23 78 Fax : 32-2-629 28 80 Email: dean@stro.vub.ac.be http : to be announced Participants Vrije Universiteit Brussel Intecs Sistemi ENEL AMS GmbH Visual Computergraphic Arakné Country B I I D IRL F Start Date 1st October 1998 Duration 30 months Octobre 1998 Role C P P P P A High Performance Computing and Networking / 159 28189 ALESSIA Application of Large Eddy Simulation to the Solution of Industrial problems Keywords : High Performance Computing and Networking, Simulation Technical : Large Eddy Simulation Uses: CFD, Acoustics Industrial Applications : Aeronuatics, Automotive, Chemical, Rotating Machinery The primary aim of the project is to develop validated and supported software tools for the prediction of industrially important fluctuating flow problems. The tools are based around the technique of Large Eddy Simulation (LES), which solves for the large scale fluctuating flows, the Large Eddies, and uses ‘sub-grid’scale turbulence models for the small-scale motion. The leading European software packages, CFX for Computational Fluid Dynamics (CFD), and SYSNOISE for the acoustics, will be coupled together in this project to perform the prediction of industrially relevant flows, and the consequent noise field. The project contains a range of industrially important applications with different flow modelling issues. The areas considered are aerodynamic noise, cyclone separators, mixing vessels, multi-phase flows in pipelines and rotating machinery. The benefits are expected to be the following: design cycle times reduced by up to several months, reduced costs of supporting in-house software developments, simulations carried at at under 15% of the cost of comparable large-scale experiments, and increased software sales for the software developers. In addition, the economical and social cost of noise pollution is enormous and even a slight noise reduction can have a major impact in terms of increased allowed traffic or reduced pollution. There are also considerable environmental benefits, by allowing equipment to operate at higher speeds, and being more efficient in separating and recycling waste materials. Contact Point Dr. Ian Jones AEA Technology Harwell 8.19 UK-OX11 ORA Didcot Tel : +44 1235 432464 Fax : +44 1235 433174 Email: ian.p.jones@aeat.co.uk Http : Participants AEA Technology ALFA-Laval CR-Fiat LMS-NIT Schlumberger Country UK S I B UK Role C P P P P Shell NL P TU-Munich D P Start Date October 1998 Duration 36 This page is located at www.cordis.lu/esprit/src/28189.htm It was last updated on 29 October 1998, and is maintained by Pertti.Jauhiainen@cec.be Octobre 1998 High Performance Computing and Networking / 160 28195 AMEGOS Automatic Mesh Generation and Optimisation for Industrial Flow Simulations Keywords : High Performance Computing and Networking, Simulation Technical : Automatic mesh generation Industrial Applications : Automotive, Electrochemical The objective of the project is to develop a generally applicable grid generation toolkit for unstructured grids, suitable for the simulation of industrial devices and processes in a wide range of industrial applications. These applications are characterised by the presence of strong convective-diffusive effects and have in common the cumulative difficulties of (1) complex internal geometries, (2) moving boundaries and (3) extremely thin diffusion layers near wetted surfaces. The resolution of these diffusion layers by the grid is crucial for the performance prediction of the industrial device or process, as they determine heat transfer and mass transfer phenomena.The technical objectives of the project are the following: 1. to develop two different methods for generating high quality unstructured grids in complex configurations, namely the frontal hybrid grid generation approach and the medial axis approach. 2. to develop grid evolution tools for moving geometries. 3. to develop suitable a priori grid quality indicators based on the solution of a model convection diffusion equation, and a posteriori grid quality indicators based on finite element residuals of the fluid or charge transport equations. 4. to test and validate the new technologies for grid generation and optimisation in two different critical industrial applications, namely piston engine design and electrochemical plating reactors. The business objectives are to reduce the time to market for the products by providing better simulation tools in the design cycle, and to the integrate the new grid generation toolkit developed within the project with existing flow or electrochemical solvers with the aim of commercialisation . Contact Point Dr. Wilhelm Brandst ätter AVL LIST GmbH Hans-List-Platz, 1 A-8020 Graz Tel : +43 316 787 313 Fax : +43 316 787 137 Email: brandstw@avl.com Http : Participants AVL LIST GmbH Von Karman Institute ELSYCA AB Volvo PSA Peugeot Citroen Country A B B S F Start Date October 1998 Duration 36 Role C P P P P This page is located at www.cordis.lu/esprit/src/28195.htm It was last updated on 29 October 1998, and is maintained by Pertti.Jauhiainen@cec.be Octobre 1998 High Performance Computing and Networking / 161 28247 VISiT Virtual Intuitive Simulation Testbed Keywords: High Performance Computing and Networking Technical - Virtual Reality, Interactive Simulation, Computational Fluid Dynamics, , Parallel Computing, Grid Generation Uses - Turbomachine, Automotive, Paper Production and Aerospace Industry Industrial applications - Hydro Turbine Components, Car Climate System, Paper Production Machine Design, Composite Aircraft Components The goal of the project is the development of the interactive, intuitive 3D-simulation environment VISiT by using Virtual-Reality techniques (VR) and HPCN tools partly developed already in other research projects. The end-users of the consortium come from different industries to ensure that the VISiT environment is applicable independent of the application field with focus on CFD. The technical objective of the project is to establish a numerical workbench, designed for intuitive, interactive evaluation of simulation and intuitive, interactive modification of boundary conditions and geometry to directly transpose the finding and the new ideas of the working team into a successive simulation. This VISiT workbench will be based on Virtual Reality Technologies as man-machine interface to interact directly with the processes and the data, but standard input/output devices will also be included in this environment. The efficiency of the development process will be improved considerably, thus leading to a strong reduction of cost and especially of time-to-market for the European industry. The external exploitation of the results of the VISiT project is done directly by the partners. Contact Point Dr. Franz R. Klimetzek Daimler-Benz AG FT1/AK - E223 D-70237 Stuttgart Participants Daimler-Benz AG British Aerospace CFD & Structural Engineering GmbH University of Stuttgart Vircinity-IT Consulting GmbH De Pretto-Escher Wyss Voith Hydro GmbH University of Jyväskylä Valmet Corporation Start Date October 1998 tel: +49 711 17 20859 fax: +49 711 17 58346 email: klimetzek@str.daimlerbenz.com http : to be announced Country D UK D Role C P P D D P A I D SF SF A A P A Duration 36 months This page is located at www.cordis.lu/esprit/src/28247.htm It was last updated on 13 October 1998, and is maintained by Max.Lemke@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 162 28249 HISTRIDE A High Performance Structural Identification Environment Keywords : High Performance Computing and Networking Technical : Finite element methods Uses : Structural identification, safety analysis of structures Industrial Applications : Construction, building restoration The objective of the project is to develop a high-performance computational environment for structural identification, and to demonstrate the exploitation potential of this novel technology for applications in the construction engineering industry. The HISTRIDE environment will incorporate facilities for data import, data analysis, physical model identification, and damage evaluation from the identified structural model. An open software architecture will be adopted to permit the introduction of new or alternative analysis components which may be required by the end-users. The HISTRIDE environment will be portable to low-cost systems such as PC and workstation clusters, as well as to scalable parallel super-computers. Contact Point Luca Bolognini ENEL - Polo Idraulico & Strutturale Via Pozzobonelli 6 I - 20162 Milan Tel : 39-02-7224 8503 Fax : 39-02-7224 8450 Email: bolognini@cris.enel.it http://www.pis.enel.it/histride Participants ENEL Finite Element Analysis Ltd. ISMES S.p.A. METRABVIB RDS University of Southampton, PAC Scott Wilson Kirkpatrick and C° Ltd. Country I UK I F UK UK Start Date 11 September 1998 Duration 30 months Octobre 1998 Role C P P P P P High Performance Computing and Networking / 163 28283 HEATMASTER Induction Heating Mastering for Metal Processing through Inverse Analysis Keywords : High Performance Computing and Networking Technical : finite element, inverse analysis, parallelisation Uses / Industrial Applications: Induction heating, pre-heating, surface treatment Summary Induction heating is one of the main heating processes used by metal processing plants. Lowfrequency induction can be used to carry out pre-heating of metal parts before deformation. Highfrequency induction is a means of carrying out heating of the outer surface of the part; as such, it is often used in surface treatment processes. HEATMASTER will provide industrial end-users with an accurate and parallel numerical modeling software which will simulate metal processing from preheating down to final surface treatment, and determine optimal induction heating parameters in respect to the final properties aimed at. In order to do so, this software will: - couple a finite element model for induction heating with one for analysis of thermomechanical deformations; - carry out inverse analysis techniques to access optimal process parameters; - use parallelisation techniques to reduce computation times. The various techniques which need to be carried out in order to complete successfully the project has led us to set up a consortium which includes laboratories dealing with numerical modeling (Cemef, FS-LNMS), induction heating specialists (EA Technology), industrial end-users (BLMaskinfabrik, SIAP-TQT, UEF Chesterfield Cylinders) and a software company (Transvalor). The results of this project will help industrial users of induction heating systems to get more out of their equipment. It will also help them in testing new technological solutions; changing some process parameters in an induction heating system can sometimes mean quite a lot of investment; in this sense, numerical modeling can help in determining whether one change or another can improve productivity. At this time, there is no other software on the market which can carry out a global optimization of process parameters in induction heating. Therefore, there are numerous market opportunities for such a software. Contact Point Dr. François BAY Tel : +33 (0)4 93 95 75 56 Fax : +33 (0)4 93 65 43 04 Email: francois.bay@cemef.cma.fr http : www-cemef.cma.fr Participants ARMINES-CEMEF BL MASKINFABRIK EA TECHNOLOGY FS-LNMS TQT TRANSVALOR UEF Chesterfield Cylinders Country F DK UK SL I F UK Start Date 01/10/98 Duration 30 months Octobre 1998 Role Co-ordinator Partner Partner Partner Partner Partner Partner High Performance Computing and Networking / 164 28350 BLOODSIM Simulation of cardiovascular and other biomedical problems Keywords: High Performance Computing and Networking Technical – CFD simulation, stress analysis, coupled problems Uses – Prostheses design, Cardiovascular modelling, Health applications, Biomedicine The project will produce a simulation tool to study the human cardiovascular system, and will advance significantly the use of simulation within the bio-medical community at a time when the use of prostheses is increasing substantially and clinicians are participating as never before in the engineering of organ replacements. The increasing availability of relatively inexpensive HighPerformance Computing and Networking (HPCN) means that these problems are, for the first time, tractable. The target end-users are primarily those clinicians, clinical scientists and manufacturers developing artificial heart valves, heart pumps, stents and grafts. As an example of the severity of the clinical problems, in recent years several mechanical heart valve designs have been investigated because of catastrophic failure in patients. These investigations have highlighted the lack of a quantitative understanding of the opening and closing motions of the disc occluder. This project will provide the necessary analytical facility and will aid the understanding of the behaviour of existing disc valves as well as providing an essential element in the design of new devices. The concepts used to obtain this computational solution will provide an opportunity to study many other problems in the cardiovascular field where blood flow interacts with implanted devices, including stents, replacement vessels and the fully-mechanical heart. The resultant analysis system will have a significant opportunity for exploitation in other, unrelated, areas with similar types of problems. Cardiovascular simulation is a coupled problem. Not only is blood an inhomogeneous, anisotropic, non-Newtonian fluid, but the boundaries of the flow (the arteries, veins, heart, etc.) are not rigid, and in many instances can have a pronounced effect on the flow which cannot therefore be predicted using rigid wall, or prescribed-boundary-motion approximations. So a multi-disciplinary approach must be adopted, and a focused consortium driven by a group of end-users, and clinical scientists, together with specialists in Stress Analysis and Computational Fluid Dynamics has come together to deliver the required technology. The approach taken is to enhance a commerciallyavailable CFD code and a leading commercial stress analysis package, by integrating complementary functions to take account of the particular problems posed by cardiovascular simulations. The project will develop a tool that will not only provide clinicians with hitherto unavailable insights into the various mechanisms involved, but also significantly aid in the design of cardiovascular prostheses generally. Additionally, it is anticipated that the regulatory authorities involved in validating prostheses will wish to use the system as a tool in issuing directives and validating the compliance of marketed prostheses. Contact Point: Dr Ian Jones AEA Technology 8.19, Harwell Street 0X11 ORA Didcot – United Kingdom Octobre 1998 tel +44 1235 432464 fax +44 1235 433174 e-mail: ian.p.jones@aeat.co.uk http: High Performance Computing and Networking / 165 Participants AEA Technoogy University of Sheffield Autogenics PFD IDAC CR BardD Angiomed ASD Mediport Country UK UK UK IRL IRL IRL D D D Start Date September 1998 Duration 36 months Role C P A A A A A A A This page is located at www.cordis.lu/esprit/src/28350.htm It was last updated on 12 October1998, and is maintained by Massimo.Luciolli@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 166 28363 MADS Multipurpose Antenna Design Simulator Keywords : High Performance Computing and Networking Technical : Electromagnetics Uses : Antenna design Industrial Applications : Air and traffic management, radars, broadcasting, telecommunications The objective of the project is to specify, develop, implement, and start the exploitation of an HPCN-based CAE system, called Multipurpose Antenna Design Simulator, and devoted to the electromagnetic engineering of active or passive antennas installed on their platforms or in their environment, to be used during the whole antenna design cycle. The result of the project will be a general and user-friendly computer-aided antenna design framework distributed over a computer network and taking advantages of HPC technology. It will address applications such as air and traffic management, radars, broadcasting and telecommunication system. The expected benefits with the use of MADS in the design process of an antenna system are: 1 - reduced design cycle time, 2 - better reusability of components or sub-systems, 3 - higher reliability of the design process, 4 - better quality in antenna production, 5 - more powerful man-machine interface, 6 - introduction of concurrent engineering environment. And consequently, reduce cost (from 10 % to 20 %) and the time to market (~30%). Contact Point Jean-Paul Martinaud Thomson-CSF / RCM 1, Boulevard Jean Moulin F - 78990 Elancourt Tel : 33-1-34 59 60 01 Fax : 33-1-34 59 70 54 Email: martinaud@rcm.thomson.fr http : to be announced Participants Thomson- CSF/RCM Ingegneria Dei Sistemi TICRA Quadrics Supercomputers World CIRMA Ecole Polytechnique Fédérale de Lausanne ICCS/NTUA Country F I DK UK I CH GR Role C P P P P P P Start Date Duration 15 October 1998 30 months This page is located at www.cordis.lu/esprit/src/28363.htm It was last updated on 13 October 1998, and is maintained by Renato.Campo@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 167 28372 HPNURSA High Performance Numerical Unsteady Rotor Stator Analysis Keywords : High Performance Computing and Networking, Simulation Technical : Unsteady rotor/stator coupling, flow simulation Uses: Computational Fluid Dynamics (CFD) Industrial Applications : Turbomachinery HPNURSA is dedicated towards the better understanding of rotating and non-rotating unsteady behaviour of turbomachinery components using computing simulation. The objective is to get full industrial parallel software versions able to simulate real unsteady coupling between rotating and non-rotating industrial components both for incompressible and compressible flows. It also includes a post-processing part, which is a key point for exploitation of the computational results. Within this project, the main industrial application is directly related to the performance increase of turbomachinery. Currently almost all design tools neglect the unsteady effects caused by the interaction between the rotor and the stator blade rows in turbo-machines. However, these effects have an important influence on the performance and the environmental impact of the final design. Especially the problem of noise generation (and the associated performance loss) cannot be handled on the basis of the current simplified analysis tools. At project completion, three major flow simulation programs will have enhanced their capabilities, which shall be directly available on the market at this time. Two major European end users in turbomachinery field will have increased their know-how and integrated these new flow simulation and post-processing capabilities in their own turbomachinery design process. At last dissemination of the results will be achieved by taking advantage of all existing communication media (publications, Web Site, European seminar). The strategic position of all partners in their fields guarantees a wide dissemination. Contact Point Mr. Alois Nichtawitz Tel : +43 70 6987 33 22 Voest Alpine Machinery Fax : +43 70 6987 23 59 Lunzerstrasse, 78 Email: nictaw@vamce.co.at A-40312 Linz Http : Country Role Participants Voest Alpine Machinery (MCE) A C HV CH P SEP F P AEA D P SIMULOG F P NUMECA B P LMH CH P ECL F P Start Date November 1998 Duration 32 This page is located at www.cordis.lu/esprit/src/28732.htm It was last updated on 29 October 1998, and is maintained by Pertti.Jauhiainen@cec.be Octobre 1998 High Performance Computing and Networking / 168 28425 BISANTE Broadband Integrated Satellite Network Traffic Evaluation Keywords: High Performance Computing and Networking, Simulation, Network Modelling Technical - Application Characteristics and Users' Behaviour Modelling - Network Modelling (Terrestrial, Mobile, GEO and LEO Satellite Networks), Traffic Evaluation Uses - Network Planning, Capacity Management, Optimisation, Impact of new Technologies Industrial Application - Experimentation and Validation of Network Simulator Prototype The ultimate aim of the information revolution is to ensure that everyone can have access to information, anytime, anywhere at reasonable cost. To achieve this goal, the emerging satellite technology gives a new perspective for a universal access to the broadband infrastructure, potentially alleviating the prohibitive cost of serving every user by terrestrial digital networks. Satellite-enabled personal communication services including two-way voice, fax, data, video, teleconferencing, multicasting and broadcasting are becoming increasingly affordable to residential, but also and more importantly, mobile users. Furthermore, by ideally complementing the terrestrial networking infrastructure and because of the possibility to integrate fast global telecommunication services into one single network, satellite communication appears as the key technology for the emerging "information society". Besides telecommunications, distributed and multimedia computing have evolved over the last few years. As opposed to conventional computing and data communications, multimedia exhibits whole new classes of system requirements with respect to capturing, storing, streaming, transmission, synchronization and presentation, commonly referred to as Quality of Service (QoS) requirements. First approaches in coping with those requirements at the network traffic level have led to the evolution of an ATM-based Broadband Integrated Services Digital Network (B-ISDN). These networks can not only support high transmission rates, but can also allow different applications or multimedia streams to be transmitted simultaneously in an integrated manner. Terrestrial and satellite links will provide ubiquitous access to a plethora of multimedia services over B-ISDN and to the Internet. Geostationary satellites and terrestrial networks will provide access to fixed residential users and companies. For mobile users, it appears that a very promising solution is the installation of Wireless Local Loops (WLL). These access methods can also be combined with other existing terrestrial networks. To efficiently handle multimedia applications across networks, and in particular through the air interface, appropriate network capacity management mechanisms are necessary. Network capacity planning is the practice of anticipating the networked users future needs, and designing and managing the network resources accordingly. Two key issues play a dominating role, first the characterization of the type and quality of service a user expects from the network, and second, the characterization of the type and quantity of load a user generates into the network. BISANTE will characterize accurately these parameters by modeling and analyzing user behavior, and by developing models of network load and traffic as induced by user activity in future multimedia applications. A network capacity planning workbench will help the European telecommunication industry to early identify complexity reducing the time-to-react to approaching telecommunication trends and hence becoming more competitive in the global broadband satellite communication market. Octobre 1998 High Performance Computing and Networking / 169 The workbench will include the future activities related to network capacity planning like sizing the network, allocating dedicated bandwidth, bounding latency, and guaranteed end-to-end QoS. Furthermore it will also allow for detailed network performance analysis like the evaluation of protocol designs, traffic shaping or the investigation of routing policies. BISANTE will develop network system and network traffic models based upon profiles of user behavior. It will establish insight on how new multimedia and network computing technologies (like e.g. Java, CORBA and VRML) will influence network performance, and maybe alter user behavior. The mapping and implications of QoS parameters at various system levels (networkperceived QoS, user-perceived QoS) will also be investigated. The means for network model evaluation are large-scale discrete event simulations. Appropriate approaches to the network simulation scalability problem is the use of parallel and distributed simulation techniques, where the simulation task is decomposed and assigned to a set of processors which execute subtasks concurrently in a coordinated way. A complementary solution is to slim down the simulation by abstracting out details under controlled risk, thus enabling modular, hierarchical simulations at arbitrary levels of accuracy. The simulation results will be compared with measurement results from a test-bed to verify the simulation models. Investigations will encompass the characteristics of Low-Earth-Orbit (LEO) and GeostationaryOrbit (GEO) satellite constellations, yielding modular satellite network models. The user behavior and application characteristic models will then be integrated in order to carry out simulations to verify that the application receives the level of service it requires while efficiently using the capacity of the air interface. The effects of the multiple access scheme (TDMA, FDMA, CDMA) and of the satellite link characteristics on the applications will be investigated to optimize the utilization of network capacity. Contact Point Dr. Rigobert FOKA Thomson-CSF Communications 160, Boulevard de Valmy F - 92704 COLOMBES France Participants Thomson-CSF Communications (TCC) Institut National des Télécommunications (INT) University of Surrey University of Vienna SOLINET NETWAY Start Date October 1998 phone : 33 1 4130 3425 fax : 33 1 4130 3560 e-mail : rigobert.foka@tcc.thomson-csf.com http:// Role Country Coordinator France Associated Partner France Associated Partner United Kingdom Associated Partner Austria Partner Germany Partner Austria Duration 24 months This page is located at www.cordis.lu/esprit/src/24950.htm It was last updated on 29 October 1998, and is maintained by Merce.Griera-I-Fisa@cec.be Octobre 1998 High Performance Computing and Networking / 170 29381 ARIS Action for Research and Information Support in civilian demining - a network of excellence to help improve the effectiveness of R&D activities in demining technologies Keywords : demining, multi-sensor data fusion, databases, technology assessment Technical : Uses : Industrial Applications : humanitarian demining The NoE provides a forum for the exchange of ideas to help resolve technological problems related to demining actions. It shall work to improve information exchange between R&D teams and users. Its scope includes user needs, equipment and technology test and evaluation reports; threat descriptions, operational experience, problems, and current R&D. The NoE will work to encourage technological innovation to match specific operational needs by promoting interaction with users through seminars on current and future R&D areas. The NoE shall run technical workshops on the topics of User Needs and Standards for Data Exchange, Testing and Validation. The NoE will be active on training, development strategies for R&D, EU-US cooperation. The NoE will co-ordinate research activity and develop collaboration between research teams and industry. Contact Point Dr. Alois Sieber Tel : +39 0332 785339 EC JRC Space Applications Institute Fax : +39 0332 785469 Via E. Fermi Email: alois.sieber@jrc.it I - 21010 ISPRA http: Participants EC JRC SAI RMA DLR Thomson TME CAT Science TNO FEL FOA RNLA University of Bari Country I B D F DK NL S NL I Start Date 1 October 1998 Duration 12 months Role C P P P P P P P P This page is located at www.cordis.lu/esprit/src/29381.htm It was last updated on 13 October 1998, and is maintained by Patrick.VANHOVE@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 171 28410 Roxy "New-Media Toolkit" for multimedia products and services demand Keywords: interactive audio-visual multimedia Technical: Internet, multicasting Uses: TV programs, video conferences, multi-user games, and World Wide Web services Industrial application: "New-Media Toolkit" for multimedia products and services demand Roxy: The Roxy objectives are to develop, deliver and exploit software tools, and a set of operational services for the creation, delivery, and presentation of interactive audio-visual multimedia information. Roxy services will be a convergence and combination of TV programs, video conferences, multi-user games, and World Wide Web services. The platform is Internet, exploiting the new essential technology of multicasting. The reason for targeting the general media area is to open a broad market of users, all the way to the home environment. The Roxy approach will start with a market and technical analysis of following application areas: multimedia magazines and distance learning. The project will then specify and develop a set of software tools, in order to form a "New-Media Toolkit" by expanding audio-visual and Internet communication functions. The New-Media Toolkit will exploit emerging and newly standardised technologies such as multicasting and resource reservation on Internet, audio-visual object-based encoding/decoding, advanced 2D/3D-presentation and multi-users' collaboration. Roxy will utilise the most recent technologies for platform independent and on-demand delivery of software, such as Java and mobile code solutions. Roxy is determined to get the results into actual use to establish business cases in a rapidly emerging market leveraging multimedia products and services demand. Relevant and high-quality feedback on innovation results will be achieved by actual use in field tests in Sweden and Germany. Contact Point: Ingemar Oja Viewrope Storgatan 9 S-972 38 LULEA SWEDEN Participant Viewrope THOMSON multimedia R&D France Telia CDT/Lulea University of Technology Bertelsmann GmbH&Co KG Siemens AG / Frauenhofer Institut für Graphische Datenverarbeitung TV4 Norrbotten Swedish Educational Broadcasting Company Start Date September 1998 Tel: +46 920 23 07 31 Fax:+46 920 23 03 61 E-mail: ingemar@viewrope.se http:www.viewrope.se http://www.roxy.nu/ Country Role Sweden C France C Sweden C Sweden C Germany C Germany C Sweden Sweden A A Duration 24 months This page is located at www.cordis.lu/esprit/src/28410.htm It was last updated on 29 October 1998, and is maintained by Merce.Griera-I-Fisa@cec.be Octobre 1998 High Performance Computing and Networking / 172 28842 KLEE&CO KNOWLEDGE AND LEARNINGS ENVIRONMENTS FOR EUROPEAN AND CREATIVE ORGANIZATIONS Keywords : Learning and Training in Industry, Technical : Knowledge management, colaborative tools, Human factors Uses : Creative design, communities and team building, Informal learning Industrial Applications: Industrial design, electronics and cloths industry Summary The main scope of KLEE & CO is to enhance the creative processes supporting the informal learning process mediated by the “knowledge of experience”, that today is only part of personal memories of individuals. KLEE & CO aims at designing a knowledge management support system for design centres, so that they can improve their learning, their performing capabilities, their relations with the supply chain. It will include a collaborative environment and a set of knowledge management tools and will support the process of creative production in organizations that manage the full development (from conception to realization) of innovative artifacts. The system will have an open architecture based on the Internet e-mail system and on the WorldWideWeb, and will develop and integrate state of the art “ Communityware” systems supporting casual participation, peripheral awareness of other's activities, and easy access to these communities that form out of shared interest and practices. This first release of the system will be experimented at two different design centres. Both the experiments will be carefully observed so that both important knowledge about the requirements of a newly conceived system and guidelines for installing it within a design centre will be achieved. Moreover, innovative interaction modes will be conceived and prototyped. Contact Point Dr. Enrico Donati / Prof. Frederico Butera Instituto RSO Piazza Giovine Italia 3 I-20123 Milano Tel : +39/02 48016162 Fax : +39/02 48016195 Email: info@irso-bep.it http : Participants Instituto RSO DOMUS ACADEMY PHILIPS UNIMI XEROX Country I I NL I UK Start Date Duration 18 months Role C P P P P This page is located at www.cordis.lu/esprit/src/28842.htm It was last updated on 29 October 1998, and is maintained by Khalil.Rouhana@cec.be Octobre 1998 High Performance Computing and Networking / 173 28928 KNOWNET KNOWLEDGE MANAGEMENT WITH INTRANET TECHNOLOGIES Keywords : Learning and Training in Industry Technical : Knowledge management, Collaborative working, Intelligent assistants Uses : knowledge and experience sharing, team building Industrial Applications : Finance and service sectors Summary The KNOWNET project aims to address the Knowledge Management (KM) needs of business entities by developing, applying, testing and evaluating: • an intranet-based tool at the enterprise-level, collaborative tools supporting communities of practice at the team-level, and an individually adaptable tool at the individual-level. KNOWNET aims at adopting a total systems solution and achieving technical innovation by integrating: ◊ the process-centred view, which focuses on Knowledge Management as a social communication process and is enabled by groupware support; within this view KNOWNET intends to extend, integrate and exploit the products and tools of one of the partners KNOWLEDGE ASSOCIATES (UK); and ◊ the product-centred view, which focuses on the creation, storage and reuse of explicit knowledge objects; within this view KNOWNET intends to extend and integrate the intelligent assistants approach of DFKI (D) another partner in the project;. • management methods, i.e. strategies, processes and structures, that enhance and facilitate organisation-wide knowledge leveraging and methods for explicitly measuring and evaluating the quality and business value of intangible assets; KNOWNET will build on related work of INSEAD (F). In order to verify and validate the applicability and usability of the resulting knowledge infrastructure, KNOWNET will test and evaluate the applicability and exploitability of results by installing and assessing the impact of KNOWNET in the improvement of learning capabilities in three (3) service-oriented business organisations. The need for industrial knowledge management is driven by the convergence of external market forces (such as speed of change, cycle-time reduction, globalization, etc.) and internal changes (like the recognition that downsizing and restructuring have resulted to considerable losses in organisational knowledge). Contact Point Dr Gregory Mentzas, Assistant Professor Institute of Computer & Communication Systems National Technical University of Athens 42, 28th October str., 10682 Athens, Greece Octobre 1998 tel: + 30 - 1 - 772 – 3895 fax: + 30 - 1 - 772 – 3550 Email: gmentzas@softlab.ntua.gr http://www.know-net.org High Performance Computing and Networking / 174 Participants Planet Knowledge Associates INSEAD DFKI ICCS FHBB SBC GW Country GR UK F D GR CH CH UK Start Date 1 October 1998 Duration 18 months Role C P P P A A A A This page is located at www.cordis.lu/esprit/src/28928.htm It was last updated on 29 October 1998, and is maintained by Khalil.Rouhana@cec.be Octobre 1998 High Performance Computing and Networking / 175 28942 MOST Medical representative On-line Simulation Training Keywords: IT for Learning and Training in Industry Technical: Simulation, Internet and Multimedia, Computer Based Training Industrial Applications: Pharmaceutical medical representatives MOST aims at providing a full Internet simulation based system dedicated to Medical Representatives. it will be composed of an authoring and a learning environment available on standard browsers through the Internet. Mathematical models simulating the drug's effects will ensure the simulation dimension. A course will be a specific composition of learning multimedia material managed by simulation results. In addition, the MOST Project plans to incorporate the concept of streaming with the simulation to address one of the most important technical issues: the narrow bandwidth available for multimedia material on Internet. As part of the training of medical representatives through simulation streaming based courses, we can reasonably expect, depending on the calculation power of the server available, to tackle more complex systems and to address nomadic computers. For preparing future extensions, MOST will also be tested by nurses and doctors. The potential to use the system as a daily commercial support for dedicated Medical Representatives will also be evaluated. This project will benefit from the successful Tap project COAST that produced numerous mathematical models dedicated to cardiology and neuro-physiology that will be made available for the MOST Projects. The MOST Project will require the following different profiles: 1. Pharmaceutical laboratories: Sanofi Winthrop(F) and Pfizer (B), which are among the major world pharmaceutical companies (Pfizer 40 000 employees, Sanofi 20 000 employees), will be Pilot sites and course providers. 2. In addition, the Salzburg Hospital (A) will have a special assignment for assessing the system in training both nurses and doctors at different level of complexity. 3. Mathematicians and simulation experts will come from: The «Laboratoire de Neurophysiologie» of UCL (B), which has already developed 20 mathematical models in cardiology and neurophysiology and the Risc Institute of the University de Linz (A), Research Institute for Symbolic Computation, dedicated to mathematics. 4. Multimedia material provider: the three pilot sites are potential providers, but the project will benefit from a specialist in virtual reality: Synkronix (UK). 5. As IT specialist on Internet, Selisa has carried out numerous successful European projects around education and electronic commerce on the Internet. 6. Marketing specialist: The general activity to prepare the market access will be coordinated by Excelsys. Contact Point Tony Greenwood SELISA Cerisaie 207- rue de Chevilly 94262 FRESNES cedex Octobre 1998 tel: +33 (0) 1 49.84.63.00 fax: +33 (0) 1 49.84.63.01 Email: most@selisa.fr http:// High Performance Computing and Networking / 176 Participants Selisa Ucl-Nefy Risc-Linz Sanofi Winthrop Pfizer Landeskrankenanstalten Salsbourg Synkronix Start Date January 1999 Country F B A F B A Uk Role C P P A P P A Duration 24 months This page is located at www.cordis.lu/esprit/src/24950.htm It was last updated on 29 October 1998, and is maintained by Agnes.Bradier@cec.be Octobre 1998 High Performance Computing and Networking / 177 29255 KLIF A Knowledge and Learning Infrastructure Keywords: IT for Learning and Training in Industry Concerted action. The overall objective of this concerted action is the co-ordination through knowledge sharing facilitation of the projects that have been selected under the thematic call "IT for Learning and Training in Industry". This includes the sharing of knowledge, results and ideas among the funded projects, between the projects and the European Commission, and between the projects and the rest of the world. In order to achieve this goal, a knowledge and learning infrastructure (KALIF) will be developed. This infrastructure will consists of two major components: a people framework and a IT framework. The people framework aims at the creation of enduring relationships between the organisations and people involved in the programme, by means of facilitated meetings. The IT framework will support the further elaboration of this people framework by providing means for exchanging knowledge, ideas, questions, etc. electronically. More concrete, KALIF aims to achieve the following objectives: 1. KALIF will take initiatives to improve the visibility of the projects, their products and the knowledge they developed to a wider (scientific, industrial and public) audience. 2. KALIF aims at optimising knowledge sharing between funded projects. That is: 2.1. prevent duplication of research efforts; 2.2. accelerate knowledge creation and distribution; 3. KALIF will help the projects to capture and capitalise on lessons learnt within the projects. 4. The networks of the KALIF consortium partners will be exploited to capture user requirements in an early stage to ensure maximal market potential for the products generated by the projects. 5. Furthermore, KALIF will undertake actions to stimulate the establishment of a broad and sustainable community of practice of innovative industrialists and researchers in the field of industrial and organisational learning in Europe. Contact Point Eelco Kruizinga Arthur van Schendelstraat 570 Postbus 19210 3501 DE Utrecht Participants CIBIT ECLO Start Date September 1998 tel: +32 30 320 89 15 fax: +32 30 230 89 99 Email: ekruizinga@cibit.nl http://www.kalif.org Country NL B Role C A Duration 30 months This page is located at www.cordis.lu/esprit/src/29255.htm It was last updated on 29 October 1998, and is maintained by Agnes.Bradier@cec.be Octobre 1998 High Performance Computing and Networking / 178 29479 MACADAM Multisensor Acquisition Campaign for Analysis and Data fusion of Antipersonnel Mines in support of ESPRIT projects Keywords : HPCN , anti-personnel landmines , support action Technical : Multisensor systems , landmines signature , data fusion, embedded systems Uses : Demining Industrial Applications : Landmines detection systems Today, there is no multisensor data base of anti-personnel landmines accessible by research and industry working in the field. There is a need to build such databases where the testing environments and the databases are open and available to all who are trying to develop solutions to the humanitarian demining problem. The Ojectives of the MACADAM project are to : è perform measurements with a set of sensor (EMI, GPR, IR and passive radiometer) on the same test ground ( using the facilities of the JRC in ISPRA) è to restore the resulting single sensor data in order create a data base to be used to look for the most promising data fusion approach applied to antipersonnel landmines detection. è to label this data with all information useful to characterise the sensor, the mine or mine-like object and the environment. è to study the necessary modifications of the data base management software already developed by TME in order to integrate in it the data gathered in ISPRA and to have a prototype of data base with qualitative assessment of each single sensor behaviour. è to disseminate this data base to an agreed list of users for research activities in the field of humanitarian demining. The main result of the MACADAM project will be a complete data base which will represent the first set of multisensor data available in Europe to support the ESPRIT initiative of R&D in humanitarian demining. Contact Point Philippe HERVY THOMSON-CSF/TME 23-27 rue Pierre Valette F-92245 MALAKOFF Participants THOMSON-CSF/TME Start Date August 1998 tel: +33 149 653 581 fax: +33 149 653 171 email: MACADAM@fr.tme.thomson-csf.com Country F Role C Duration 9 Months This page is located at www.cordis.lu/esprit/src/294794.htm It was last updated on 7 October 1998, and is maintained by Patrick.Vanhove@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 179 29732 CONTENTS Component Technology exercised on NT systems Keywords: High Performance Computing and Networking Technical – Advanced Visualisation, ActiveX/DCOM, CORBA, Component technology Uses – Simulation, Data Mining, Medical Imaging The objective of the CONTENTS Project is to develop customisable Application Components for the interactive visual analysis of data on powerful PC platforms. The data of interest are large multivariate data sets (engineering, medical and commercial) generated by simulation / modelling / measurement running on HPC platforms in a distributed and heterogeneous environment (including NT-based Intel parallel processor and suitable UNIX platforms). The components will be based on the emerging industry standard ActiveX/DCOM architecture, and will be properly "bridged" towards CORBA based architectures, for multi-platform portability and compatibility. These Interactive Visualisation Components will be embedded within electronic documents, allowing the author of a report to distribute the relevant electronic information coupled with an embedded High Performance data analysis-viewer "smartdoc", which allows the recipients to interactively examine the data in the same way as the original analyst. Instead of being a dumb document, the "smartdoc" will be a complete data exploration application through which a reader can, via embedded High Performance visualisation / analysis components, explore the data underlying the report. Industrial partners British Aerospace and Unilever, and the medical research hospital DFC of the University of Florence, will drive the project with their pragmatic practical user needs. Leading edge industrial SMEs AVS, AET and Intecs will provide their high tech competence and experience in data reduction and visualisation, image processing and management, multi-platform components design and re-engineering. The project will enable SME technology providers to investigate the potential of providing their tools as distributed components to supply baskets of techniques for particular applications. Contact Point: Dr Mikael Jern Advanced Visual Systems 15, Blokken 3460 Birkeröd - Denmark Participants AVS-UNIRAS AET Intecs Sistemi Unilever British Aerospace DFC-University of Firenze Start Date January 1999 Octobre 1998 tel +45 45999599 fax +45 45999500 e-mail: mikael@avs.com http: www.avs.com Country DK I I UK UK I Role C P P A A A Duration 24 months High Performance Computing and Networking / 180 29737 HPGIN High Performance Gigabit I2O Networking Software Keywords: High Performance Computing and Networking Technical: Communication for Gigabit LAN Technologies, Network Operating Systems Uses: Network Interface Controllers, Driver Standardisation, resource sharing Industrial Applications: High-performance Servers, Real-Time Applications The goal of HPGIN is to develop the Communication Layers for High Performance I2O solutions optimised for Gigabit LAN technologies. I20 (Intelligent I/O) is an International Standard that has been defined by the I2O Special Interest Group (I2O SIG) which was formed by a group of leading enterprise computing vendors to define and promote a technical specification as a standard interface for high-performance I/O systems. I2O will be the basis for driver standardisation, system performance enhancements, resource sharing, clustering and distributed heterogeneous systems. I2O will be supported by the Standard Network Operating Systems as well as by Real-Time Operating Systems. The HPGIN project includes implementation of Communication layers for a Gigabit LAN I/O system, a Standard Operating System (Linux) and a Real Time Operating System (Virtuoso). Contact Point Frank Hohmann SysKonnect Siemensstrasse 23 D-76275 Etltlingen Tel : +49 7243 502 317 Fax : +49 7243 502 989 Email: fhohmann@skd.de http : http://www.syskonnect.de Participants SysKonnect Eonic Systems University of Helsinki Country D B SF Start Date Contract unsigned at time of printing Duration 18 months Role C P P This page is located at www.cordis.lu/esprit/src/29737.htm It was last updated on 29 October 1998, and is maintained by Roman.Tirler@cec.be Octobre 1998 High Performance Computing and Networking / 181 29812 LOTUS Light Ordnance detection by Teleoperated Unmanned System Keywords: High Performance Computing and Networking, Humanitarian Demining, Technical - Multi-sensor Data Fusion, Tele-operated Vehicle Uses - Survey of Minefields for Clearance, Quality Assurance for Mine Clearance Industrial Applications - Humanitarian Demining The objective of the LOTUS project is to develop, integrate and demonstrate a proof of concept of a multi sensor APL detection system integrated on a robotic remote controlled vehicle. The remote control will radically improve safety of demining operations while vehicle based multi sensor detection combined with powerful data fusion will lead to more productive humanitarian mine detection operations. The sensors are Multi-Spectral and Multi-Dimensional: Ground Penetrating Radar, Infra Red, Metal detector, and have been studied in the on-going GEODE project. They will be further improved and adapted to a remote controlled vehicle as well as the data fusion and the computer architecture, to handle efficient real time operations. The LOTUS team has contacts with an NGO end-user and with a demining expert. These will be involved from the outset to make sure the operational requirements constraints are taken into account. They will also be involved in the preparation and implementation of experiments of the system in the field. Contact Point Jean-Philippe Quin Dassault Electronique 55 quai Marcel Dassault F-92214 Saint-Cloud Participants Dassault Electronique SA Emrad Ltd Institut Dr. Foerster Marconi SpA TNO-FEL Start Date January 1999 tel: +33 1.3481.3286 fax: +33 1. 3481.3104 email: Country F UK D I NL Role C A P P P Duration 36 Months This page is located at www.cordis.lu/esprit/src/29812.htm It was last updated on 29 October 1998, and is maintained by Patrick.VanHove@cec.be Octobre 1998 High Performance Computing and Networking / 182 29870 HOPE Hand-held Operational Demining System Keywords : High Performance Computing and Networking, De-mining Technical : Data fusion Uses: Ground Penetrating Radar, Metal detector, Microwave radiometer Industrial Applications : Humanitarian de-mining The objectives of the HOPE project is to develop and build an efficient handheld demining tool as a remarkable contribution to the European and world wide Humanitarian Demining Programs particularly in countries facing the danger caused by millions of anti-personnel mines left behind from former wars. The result of the HOPE-Project will be a handheld mine detection equipment that must: • be built with the lowest possible cost (less than 12,000 ECU) • be able to detect plastic and minimum metal mines for minimum metal mines: more than 30% deeper than today • be reliable by achieving a clearance result near 100 % • be operational in the different countries under difficult environmental conditions maximum weight: 7.5 kg • take into account the existing demining procedures • speed up the world wide demining process by more than 30 % • reduce the false alarm rate by more than 50 % These requirements can only be reached by using a multi-sensor system set: a metal detector (MD) for the indication of metallic objects, a ground penetrating radar (GPR) for the indication of metallic and dielectric objects and a microwave radiometer (MWR) primarily for the indication of flushed mines. The project will push the capabilities of the involved European Small and Medium Size Enter prises (SMEs) forward. The competitiveness of the industrial partners in the world market is increased and can keep up with parallel development programs in the USA with the same objectives. The industrial partners have planned to build up a series production after a successful sensor development. Contact Point Mr. Michael Rothe Vallon GmbH Im Grund, 3 D-72800 Eningen Octobre 1998 Tel : +49 7121 9855-31 Fax : +49 7121 83643 Email: Vallon-GmbH@t-online.de Http : High Performance Computing and Networking / 183 Participants Vallon GmbH BATS DLR ISL MAG NPA Country D B D F UK N Role C A P A P P Onera F A POLIMI I A RMA/SIC B P RST AG CH A RST GmbH D P Spacebel B P RUB D A Uni. Karsruhe (IHE) D A Start Date January 1999 Duration 24 This page is located at www.cordis.lu/esprit/src/29870.htm It was last updated on 29 October 1998, and is maintained by Pertti.Jauhiainen@cec.be Octobre 1998 High Performance Computing and Networking / 184 29895 PICE Pinpoint, Identification Clearanceand and Ensurance Keywords : High Performance Computing and Networking, De-mining Technical : Data fusion Uses: Metal Detectors (MD), Ground Penetrating Radar (GPR) Industrial Applications : Humanitarian de-mining The PICE project focuses on the realisation of a safe and cost-effective man-portable multi-sensor anti-personnel mine detection system, based on a Metal Detector (MD) in combination with a Ground Penetrating Radar (GPR). The PICE project intends and addresses the following specific objectives: - - - - Develop and verify practical and useful detection systems, targeted to established user requirements, Design a ground penetration radar with a wide band output signal and capable of classifying objects in the ground. The used technologies have been chosen with low cost production in mind. Design and validate a metal detector as a single sensor using an innovative technology with two integrated detection solutions. Design and validate a safe and cost-effective man-portable multi-sensor APL detector demonstrator comprising of the ground penetration radar and the metal detector. The sensor Human Machine Interface (HMI) must be kept simple and allow safe operation without extensive training. The demonstrators will be verified by JRC against JRC scenarios/criteria. Develop specialised software to simulate the radar signal propagation in the ground to maintain safety requirements and optimise the GPR design. Organise and "intelligently" exploit all available information on landmines, as well as expert’s knowledge on mines, through the development of an engineering database/ knowledge base. Prepare all plans and studies for adaptation of the sensors to production. Study the sensor requirements for possible use on Air-borne or vehicle-based platforms suitable for demining missions. Study the feasibility of using of an explosive sensor in the final multi-sensor product. The exploitation strategy aims to adapt the PICE system for production and investigate the potential of other technologies in relation with the long-term humanitarian-demining goals. Contact Point Mr. Claes Lindskog Tel : +46 8 580 854 09 Celsiustech Electronics AB Fax : +46 8 580 322 44 Nettovägen 6 Email: cali@celsiustech.se S-175 88 Järfälla Http : Participants Country Role Celsius S C Celsiustech Electronics AB (CTE) S P ATM Group S A FOA S P SWEDEC S A SINTEF N P NPA N A NTUA GR A Schiebel A P Start Date January 1999 Duration 24 This page is located at www.cordis.lu/esprit/src/29895.htm It was last updated on 29 October 1998, and is maintained by Pertti.Jauhiainen@cec.be Octobre 1998 High Performance Computing and Networking / 185 29902 DEMINE Improved cost-efficient surface penetrating radar detector with system on chip solution for humanitarian demining Keywords : High Performance Computing and Networking, De-mining Technical : Surface Penetrating Radar (SPR) Uses: Close-in detection Industrial Applications : Humanitarian de-mining The Project will improve existing Anti-personnel landmine (APL) surface penetrating radar (SPR) detectors in terms of detection/false-alarm rates, cost and weight. The main technical objectives for the improvements are: 1) Revolutionary system on chip solution based on high speed digital technology 2) Antenna array for multi-static and multi-polarisation techniques 3) Dynamic positioning measurement system 4) Multi-dimensional signal processing and classification which exploits the novel features of the radar The business objectives have been identified by the industrial partners and also the research institutions, aware of their economic and competitive role for EU industry in the transition to the information society: A) APL SPR prototype product which is practicable, affordable, valid and in demand B) Sub-systems and Know how products which are practicable, affordable, valid and in demand for transfer into other areas of industry C) To develop the joint-venture, investment and placement plans for A) and B) D) To ensure effective dissemination throughout the EU for the optimal exploitation of A) and B) to enhance European industrial competitiveness as a whole. Contact Point Dr. Jürgen Sachs Technische Universität Ilmenau PF 100565 D-98684 Ilmenau Tel : +49 3677 69 2623 Fax : +49 3677 69 1113 Email: sac@e-technik.tu-ilmenau.de Http : Participants Technische Universität Ilmenau MEODAT VUB AEL DERA IDS Country D D B UK UK I Role C P P P P P MgM D P Start Date January 1999 Duration 24 This page is located at www.cordis.lu/esprit/src/29902.htm It was last updated on 29 October 1998, and is maintained by Pertti.Jauhiainen@cec.be Octobre 1998 High Performance Computing and Networking / 186 29907 NEPHEW Network of PCs Heterogeneous Windows-NT Engineering Toolset Keywords : High Performance Computing and Networking Technical : Windows, NT, Clustering, PeakWare, SCI Uses : Software engineering, application development for low cost HPCN platforms Industrial Applications : Aircraft flight control simulator, film restoration, medical image reconstruction Thanks to standardisation of techniques, architectural concept and to cost reduction, HPCN has started to leave its niche market to become a more common technology. ESPRIT through initiatives like Europort and the HPCN TTN Network strongly contributed to this evolution. To really take up, HPCN must however still lower its entry cost by • relying on the most common and cost effective hardware and software found on the market place • providing easy to use tools to develop and deploy applications on these environments • being compatible with market standard development and execution environments Today, existing clustering hardware allows getting HPC performance at low hardware cost. The challenge is now to get applications for such platforms, with software engineering and deployment efforts low enough to offer attractive cost/benefit ratio to users together with a return on investment for software developers. The NEPHEW project aims at addressing this issue by implementing an environment for the rapid and easy development and deployment of professional high performance parallel applications on Heterogeneous Networks of Windows NT PCs, exploiting existing and emerging standards. At the application design level, this parallel software engineering toolset will reuse and extend the PeakWare environment originally developed by MS&I for the design and deployment of parallel real-time applications for heterogeneous target platforms. The NEPHEW environment will allow its users to express their applications as diagrams using software modules interconnected by data flow connections and, independently, target configurations described with graphical building blocks. NEPHEW will then automatically create optimised source code for communication and synchronisation. NEPHEW will also ensure interoperability with other PeakWare supports (VME, workstations, etc.). The new emerging modelling standard UML will also be introduced where suitable to facilitate application design and reuse, to exploit other tools (modelling, debugging, visualisation, etc.) users are familiar with, and consequently to facilitate future market penetration. At the deployment level, to deliver HPC performances on PC platforms, the toolset will rely on DOLPHIN SCI clustering technology and on new standards that are under adoption in the PC world to support efficient clustering of PC servers (VIA, Winsock 2). The NEPHEW environment will be validated and assessed through the development of three parallel applications in areas where HPCN is introduced as a novelty when using already existing PC based infrastructures: • The first one concerns the implementation of a real-time aircraft flight control simulation system over a cluster of PCs, with a high clock rate (simulations at 60Hz and more) replacing expensive SGI systems. • The second one concerns film manipulation modules which cannot be considered for manual MPIparallelisation because of a too low effort/benefit ratio. These density-correction modules are part of a film restoration software, for which the best effort/benefit modules have already been parallelised in the ESPRIT project FRAME 24220 by using a manual MPI-development. • The third one is the parallelisation of complex medical image reconstruction (Positron-EmissionTomography – PET) for use in hospital operations such as, for example, epilepsy care. Octobre 1998 High Performance Computing and Networking / 187 The NEPHEW environment will then be turned into a commercial off-the-shelf product to be marketed world-wide by the project partners and their commercial distributors. This environment will address the market of applications with high processing requirements at new low cost standards. This includes very large markets such as video processing, quality control or real timesimulation. The NEPHEW project will be co-ordinated by MS&I, a French company developing and marketing complex computing systems, including the PeakWare software engineering product. DOLPHIN a Norwegian SME developer and SCI’s world-wide market leader, will provide a high performance clustering solution. ELCO, a Spanish SME, will develop the aircraft flight control simulation. JRS, an Austrian business company performing applied and research development, will parallelise the library for film restoration. TUM, Germany, will develop the medical imaging reconstruction application. In summary, the NEPHEW project aims at breaking the application software costs barrier, initially for market segments such as the traditional PeakWare market, so as to leverage the availability of low-cost clustering hardware solutions. Contact Point Alain Clouard MS&I Rue Dewoitine F-78142 Velizy-Villacoublay Cedex Tel : +33 1 34 63 72 42 Fax : +33 1 34 63 74 83 Email: clouard@matra-ms2i.fr http : Participants Matra Systemes et Information Dolphin ELCO JRS Technical University Munich Country F N E A D Start Date 1 Jan 1999 Duration 18 months Role C P P P P This page is located at www.cordis.lu/esprit/src/29812.htm It was last updated on 29 October 1998, and is maintained by Tom.Clausen@cec.be Octobre 1998 High Performance Computing and Networking / 188 29909 Metodis Metacomputing Tools for Distributed Systems Keywords: High Performance Computing and Networking Technical: Distributed Computing, ATM-API, MPI Uses: Numerical Simulation, High-Speed Networking Industrial Applications: Aerospace and Automotive Engineering The central objective of the METODIS project is to develop tools that help to exploit the metacomputing capabilities that have become available on a European scale as a result of establishing European high-speed networks and supercomputing centres. In previous projects it has become obvious that optimum usage of network resources and provision of a flexible programming model are the key factors for metacomputing performance. Therefore, METODIS concentrates on these two issues. A metacomputing MPI library implemented both on TCP/IP and on ATM will serve as the application programming model. The provision of an ATM–layer will allow to flexibly and efficiently exploit networking resources, while falling back on TCP/IP will allow achieving portability in cases where ATM connectivity is not fully available. The extension of the well– established Vampir performance–analysis tool to metacomputing will greatly assist end–users in leveraging the network performance for their applications. Developing and integrating these three components will create a metacomputing environment. The ultimate measure of the METODIS achievements will be the marketing of the METODIS tools amongst academic and industrial end–users. Contact Point Karl Solchenbach Pallas GmbH Hermuehlheimer Strasse 10 D-50321 Bruehl Tel : +49 2232 1896 14 Fax : +49 2232 1896 29 Email: solchenbach@pallas.de http : http://www.pallas.de Participants Pallas CRIHAN RUS DASA Aerospatiale Country D F D D F Start Date Contract not signed at time of printing Duration 24 months Role C P P P P This page is located at www.cordis.lu/esprit/src/29909.htm It was last updated on 29 October 1998, and is maintained by Roman.Tirler@cec.be Octobre 1998 High Performance Computing and Networking / 189 29944 INFIELD Industry NGO Field test of Improved Efficiency multi-sensor Landmine Detector Keywords: High Performance Computing and Networking, Humanitarian Demining Technical - Multi-sensor Data Fusion, Human Factors, Field Testing Uses - Close-in detection Industrial Applications - Humanitarian Demining This project addresses the pressing need for a handheld demining equipment to detect plastic mines with adequate detection and false-alarm rates. It is planned to achieve this objective through the combination of Electro-Magnetic Induction (EMI), Passive Radio-Frequency (PRF) and Ground Probing Radar (GPR) sensors. The consortium is currently developing a multisensor device with the aim of fielding a product within two years. The proposed programme of work will provide the hard performance data in the field necessary to demonstrate and refine operating conditions. The technical objectives of the proposed programme are : • to evaluate the performance of prototype multi-sensor mine detector equipment, in controlled, and realistic minefield conditions, including with real deminers. • to maximise the performance of the equipment through optimised signal processing and operational procedures • to identify the most effective ways of exploiting the equipment within the demining organisational structure, and the associated cost benefits • to obtain independent performance assessment (benchmarking) The equipment will be subjected to independent assessment at the JRC, Ispra and to field tests in Bosnia and in a second infested country such as Angola. During the trials the consortium will work with end-users to investigate the most effective way of integrating the equipment into demining procedures. Trials at the JRC, in ISPRA, will allow the JRC to carry out scientific comparison of the performance of the multi-sensor handheld equipment with existing and (by establishing an appropriate database) future equipment. Contact Point François NIVELLE Thomson-CSF / TME / MDM 23-27 rue Pierre Valette F-92245 Malakoff Cedex Participants TME SA ERA TME Ltd NPA Start Date January 1999 tel: +33 149 653 161 fax: +33 149 653 636 Email: nivelle@tme.thomson.fr Country F D UK N Role C P P A Duration 18 Months This page is located at www.cordis.lu/esprit/src/29944.htm It was last updated on 29 October 1998, and is maintained by Patrick.VanHove@cec.be Octobre 1998 High Performance Computing and Networking / 190 29958 MINESEYE Development and Optimization of a dual sensor system with real time digital signal processing for the detection and identification of buried landmines and unexploded Ordnance Keywords: High-Performance Computing and Networking, Humanitarian Demining. Technical - Bulk Explosive Detection, Pulsed Neutron Systems, Chemical Analysis. Uses - Close-in detection and confirmation of landmines. Industrial Applications - Humanitarian Demining, Airport luggage inspection. The objective of the project is to develop an improved mine detection unit for Humanitarian Demining operations. The system will be based on the integration of 2 sensors, an improved metal detector and a chemical analysis sensor. The performance of the component systems will be investigated in portable scenario and on a mobile vehicle platform. The MINESEYE system will provide rapid search with a Digital Induction Pulse Sensor (DIPS) to locate a buried metallic object among various types of soils, and a Pulse Neutron Chemical Analysis Sensor (PNCAS) to detect and confirm the presence or absence of explosive substances at the location in question. This combination will substantially reduce false-alarm rates for the overall system. The work is divided into 3 phases: • Component development on the DIPS and the PNCAS units, • System integration to develop a man-portable unit based on these 2 sensors, followed by a vehicle-mounted semi-automated scanning unit, • Testing and optimization of the man-portable system in the laboratory and in JRC, together with field trials of both systems in Angola. It is expected that it will take 1 minute or less for the MINESEYE system to carry out the detection and confirmation operations. Current procedures to unearth a suspected object for inspection takes about 15-20 minutes. The saving in time using the 2 sensors Scan and Detect methodology can therefore be estimated to be at least a factor of 10. Contact Point Carmen Dumitrescu EPPRA sarl Ecole Polytechnique, Bat. X-Pole F-91128 PalaiseauCedex Participants EPPRA X-TECH JCS NV MgM Start Date Contract not yet signed tel: +33 169 33 30 98 fax: +33 169 33 43 83 Email: carmen@eppra.polytechnique.fr Country F F UK RO D Role C P P P P Duration 30 Months This page is located at www.cordis.lu/esprit/src/29958.htm It was last updated on 29 October 1998, and is maintained by Patrick.VanHove@cec.be Octobre 1998 High Performance Computing and Networking / 191 85386 PROMISE Demonstration and assessment of HPCN in medical imaging Keywords : High Performance Computing and Networking Technical : 3D Finite Element analysis, CT scan, SPECT, 3D echography Uses : remote image processing, custom prostheses design, diagnosis Industrial Applications : medical imaging applications, simulation New functionalities and services for medical imaging applications are enabled by HPCN technologies. For example 3D and 4D visualisation, image fusion from different sources, image transmission between equipment and over networks, virtual reality. As inter-operability of image handling and transmission equipment becomes possible using emerging standards such as DICOM, and as IT equipment and communication costs become more affordable, it is possible to significantly improve the quality of medical services available for the citizen. To foster the take-up of HPCN technologies in medical imaging applications the project has: – performed a survey to identify requirements of applications which may be satisfied by the use of current HPCN technologies, or which require further research; – set-up three demonstrators which were given to the HPCN Technology Transfer Nodes for wider awareness and dissemination. The demonstrator applications selected were: – 3D modelling of human bones, for surgical planning and simulation; – reconstruction of SPECT images; – 3D post-processing of ecographic images. Contact Point Dr Sanzio Bassini CINECA Via Magnanelli 6/3 I-40033 Casalecchio di Reno (Bologna) Tel : +39 051 6171514 Fax : +39 051 6132198 Email: Bassini@cineca.it Http://www.cineca.it/visit/promise/index.html Participants CINECA Istituti Ortopedici Rizzoli Dip. Fisiopatologia Clinica–Univ of Firenze University of Bologna Country I I I I Start Date January 1997 Duration 18 months Octobre 1998 Role C Sub-contractor Sub-contractor Sub-contractor High Performance Computing and Networking / 192 85387 RAIN Demonstration and assessment of HPCN in neural network applications for industry and medicine Keywords : High Performance Computing and Networking Technical : Neural Networks, workstation clusters Uses : image processing, pattern matching, forecasting and prediction Industrial Applications : medical applications, visual inspection Neural networks (NN) are increasingly used in a wide range of applications. Surveillance applications, the monitoring of scientific experiments, pattern recognition and visual inspection in quality control applications, are some examples where neural networks are used successfully to perform image processing or data handling in a range of industrial fields and in medicine. To foster the take-up of HPCN technologies in neural network applications the project has: – performed a survey to identify requirements of neural network applications for industry and medicine which may be satisfied by the use of HPCN technologies; – set-up demonstrators for industrial and medical applications, which were given to the HPCN Technology Transfer Nodes for wider awareness and dissemination. The demonstrators were set-up to run on a scalable configuration of PCs and workstations, clustered through Ethernet and ATM connections, and accessible to users from the web. The demonstrator applications selected were: – Optical Character Recognition (OCR); – Pre-press image colour correction; – Eating disorder diagnosis; – Lhyme Borreliosis disease diagnosis. Contact Point Prof Giancarlo Parodi Dept Biophysical and Electronic Engineering Via Opera Pia 11A I-16145 Tel : +39 010 3532760 Fax : +39 010 3532175 Email: gian@dibe.unige.it Http://dibe.unige.it Participants Dibe – University of Genova Country I Start Date January 1997 Duration 18 months Octobre 1998 Role C High Performance Computing and Networking / 193 85388 Demonstration and assessment of HPCN tools for mesh generation and mesh partitioning Keywords : High Performance Computing and Networking Technical : Mesh generation, mesh partitioning Uses : Demonstration, tool assessment Industrial Applications : Powder compaction, metal forming The first objective of the project was to perform a survey of existing tools for mesh generation and mesh partitioning for parallel finite differences, finite volumes and finite elements computations. Particular emphasis has been put on the analysis of their applicability in the light of the standard and innovative database oriented computational environments. This includes the assessment of the general suitability of mesh generation and mesh partitioning tools for compliance with advanced data base management systems, CAD systems, analysis codes, error estimation, mesh adaptation techniques and visualisation tools. A subset of tools has been chosen to demonstrate the possibilities of HPCN through the whole chain of pre-processing, adaptive analysis and visualisation in selected industrial problems. Three demonstrators have been developed aiming at the solution of the following industrial problems: - powder compaction - metal forming - a large scale fluid-dynamics problem Contact Point Mr. Gabriel Bugeda CIMNE Edifici C-1, Campus Norte-UPC Gran Capitan, s/n E - 08034 Barcelona Tel : +34-93-205 70 16 Fax : +34-93-401 65 17 Email: bugeda@etseccpb.upc.es Http : Participants CIMNE, Universitat Politecnica de Catalunya Country E Start Date 2 December 1996 Duration 25 months Octobre 1998 Role C High Performance Computing and Networking / 194 85389 Demonstration and assessment of scalable low-cost HPCN systems for simulation applications in non-traditional industrial sectors Keywords : High Performance Computing and Networking Technical : Simulation, workstation clusters Uses : modelling, design Industrial Applications : medical, environment, chemistry, components, construction Traditionally HPCN has been used by a small set of government departments, agencies and industries involved in modelling and simulation. In particular, for organisations in the automotive, aerospace and oil and gas sectors, the issues, benefits and cost-effectiveness of HPC are clear. Many of these are now moving away from the traditional vector parallel systems onto smaller symmetric multiprocessor (SMP) and massively parallel platforms (MPPs). As the cost of hardware has steadily fallen, the increased capacity, capability and improved competitiveness offered when using HPCN systems can be exploited by a broadening spectrum of industries. It is natural that the mature simulation technologies provided one of the first mechanisms for low-cost HPCN to be accepted and applied in the non-traditional sectors. In Particular the availability of simulation applications for clusters of workstations and PCs has meant that the entry cost for industry investing in simulation has been drastically reduced. Understanding the status and issues for this emerging market is important for the success of many European industries. The motivation for the current work is therefore the need to: - identify the new and emerging industrial sectors which are deploying or considering deploying low-cost HPCN systems in simulation applications; - determine the technical and commercial requirements of these new and potential users and the salient issues in their application domains; - demonstrate to the wider industrial community the applicability and success of scalable low-cost HPCN systems in simulation applications. The demonstrator applications selected were: – Powder compaction; – The virtual hand (prosthetic fabrication); – Traffic simulation; Contact Point Owen Thomas Smith Group Surrey Research Park UK-Guildford GU2 5YP Tel : +44 1483 442118 Fax : +44 1483 442304 Email: ogmthomas@smithgroup.co.uk Http://www.smithgroup.co.uk Participants Smith Group Country UK Start Date January 1997 Duration 18 months Octobre 1998 Role C High Performance Computing and Networking / 195 85390 Demonstration and assessment of HPCN for information management and decision support applications Keywords : High Performance Computing and Networking Technical : Information Management & Decision Support Uses : All Industrial Applications : All Growing competitive pressure implies ever more demanding requirements on applications for information management and decision support (IMDS). The size of data warehouses increases and databases searched become distributed. The response times required to remain competitive become closer to real-time, as replies to customers may be required during a phone call. The search for data becomes more complex, and sophisticated data analysis tools including e.g. data mining and optimisation algorithms, appear. To help accelerate the take-up of HPCN in IMDS applications in traditional sectors (e.g. banks and insurance companies) and to foster the take-up in new industrial and commercial sectors (e.g. retailer and telecommunication companies) the tenderers have: • identified requirements of applications for IMDS: - which may be satisfied by the use of existing HPCN applications and technologies; - which require further R&D effort supported at a European level. • promoted existing results by setting up two different demonstrator applications showing the level of maturity attained and the benefits gained through the use of HPCN applications and technologies in IMDS ; the first was a retail data mining application ; the second was a fraud detection application • proposed actions which are envisageable at a European level so that the HPCN work programme may contribute to improve the competitiveness of European industry using or producing IMDS applications Contact Point Alan Pitman The Smith Group Guildford Surrey GU2 5YP, England Participants The Smith Group University of Southampton - PAC Start Date January 1997 Octobre 1998 tel: +44 1483 442000 fax: +44 1483 442280 email: aapitman@smithsys.co.uk Country UK UK Role C P Duration 20 months High Performance Computing and Networking / 196 85391 VR FOR EUROPE Keywords : High Performance Computing and Networking Technical : Virtual Reality Uses : Engineering Design Review, Operator Training, Sales Support Industrial Applications : Aerospace, Automotive, Consumer, Virtual Reality holds the promise of a wide practical use thanks to progresses in realistic real-time computer graphics, in various type of input devices, in affordable high performance computing architectures and in an emerging understanding of the activities where VR brings a substantial added value in industry and services. The central aim of VR for Europe, is to foster the take up of VR in European Industry and Services. The following actions contribute to this aim. 1. A survey of the State of the Art. Performed with VR software, hardware or peripheral vendors and with researchers in VR or related fields (Image Synthesis, Real Time graphic simulation ...) 2. A survey of user needs and requirements Adressing industrial users at all stages of maturity in the use of VR technology. 3. Implementation of three prototypes of VR. Three industrial sites have been selected among those visited, for which VR prototypes have been developed to solve real indusrial user needs. The results of the project will be disseminated through various channels including conferences (ISTC, Vienna 1998, Imagina Monte Carlo 1999) and Technology Transfer Nodes (TTN). In particular, the somr of the demonstrators have been implemented in the CAVE environments at the Dutch TTN and TTN-PDC (Sweden) Contact Point Pierre Bouchon CCMS Dept. Avenue Pierre Brossolette 66 FR – 92247 PARIS Tel : +33 1 41 48 03 61 Fax : +33 1 41 48 06 81 Email: Pierre.Bouchon@syseca.thomson.fr Http : Participants SYSECA SA Country FR Start Date January 1997 Duration 20 months Octobre 1998 Role C High Performance Computing and Networking / 197 85392 Demonstration and assessment of HPCN in multi-site applications for virtual reality Keywords : Technical : Uses : Industrial Applications : Summary Virtual Reality integrated with High Performance Computing and Networking promises exciting new enhancements to Business processes, Science and Education. The JRC has demonstrated the importance of this field and provided this emerging technology through careful analysis and application development. Three different demonstrators, all suitable for public demonstration, are developed and transferred the HPCN TTNS. These span from Interactive Flights over a Virtual Europe – through Tele-medicine – to Virtual Presence in remote or hazardous places. From the planned survey and demonstrator development, follow-up activities and contributions to a strategy for Virtual Reality and HPCN have been suggested for inclusion in the IST programme. Contact Point Dr C.H. Best Institute for systems, Informatics and Safety Joint Research Center 21020 Ispra, Italy Tel : +39 332 785044 Fax : +39 332 789185 Email: Clive.best@jrc.it http://VR.jrc.it Participants JRC ISIS Country I I Start Date Decmber 1996 Duration 18 Month Octobre 1998 Role P P High Performance Computing and Networking / 198 501031 MINETEST Support activities to Esprit R&D projects relating to Humanitarian Demining Keywords : High Performance Computing and Networking, Humanitarian Demining Technical : Equipment validation, Data collection, GIS Uses : Testing of mine detection systems, Industrial Applications : Humanitarian Demining The European Commission (EC) supports humanitarian demining operations in various parts of the world. Current techniques for mine clearing operations are slow and costly. The EC therefore intends to assist the rapid development and introduction into service of new advanced systems. ESPRIT projects on humanitarian demining were started early 1998 and early 1999. The objective of this action is to ESPRIT projects on R & D in Humanitarian Demining and other organisations co-operating with the EC in humanitarian demining activities, by providing a series of facilities and services detailed below. Tasks 1. Setup of independent test facilities for the verification, validation and benchmarking of systems and equipment for the detection and identification of APLs. 2. Provision of access to these facilities for measurement and test campaigns to be performed by organisations selected by the European Commission. 3. Technical support to prepare and operate measurement and test campaigns. 4. Organisation of technical workshops to foster, at an international level, the communication and collaboration of experts, and the contribution to common guidelines and procedures. Contact Point Dr. John Dean Space Applications Institute CCR I – 21020 ISPRA Tel : +39 0 332 78 9407 Fax : +39 0 332 78 5469 Email: John.Dean@jrc.it Http Participant JRC Country I Start Date January 1998 Duration 24 months Octobre 1998 Role C High Performance Computing and Networking / 199 501032 MINESIGN Measurement campaign in support of Humanitarian Demining Keywords : High Performance Computing and Networking, Humanitarian Demining Technical : Data collection,Electromagnetic Signatures Uses : Training of mine detection equipment, Validation of mine detection equipment Industrial Applications : Humanitarian Demining The European Commission (EC) supports humanitarian demining operations in various parts of the world. Current techniques for mine clearing operations are slow and costly. The EC therefore intends to assist the rapid development and introduction into service of new advanced systems. ESPRIT projects on humanitarian demining were started early 1998 and early 1999. Measurement campaigns are performed to gather harmonised sets of reference data to support the ESPRIT projects in the development and validation of equipment for localisation and identification of APL’s, and to create a reference set of data for international use. Tasks 1. Executing measurements on the characteristics of APL, potential “false alarms” and background, developing and making available a library of APL signatures. 2. Collecting inputs into and giving access to a publicly available database of information on APL. Contact Point Dr. John Dean Space Applications Institute CCR I – 21020 ISPRA Tel : +39 0 332 78 9407 Fax : +39 0 332 78 5469 Email: John.Dean@jrc.it Http Participants EC Joint Research Centre Country I Start Date January 1998 Duration 15 months Octobre 1998 Role C High Performance Computing and Networking / 200 501033 SOFT-IT Software interoperability and platform independence: the next generation of simulation environments Keywords : High Performance Computing and Networking Technical : Interoperability, simulation environments, CORBA Uses : Tool assessment Industrial Applications : Electromagnetic simulation The objective of this project is to demonstrate the competitive advantages of software interoperability and platform independence in simulation environments in terms of software maintenance and exploitation potential. An analysis of software results of recently completed HPCN projects has been performed, with special emphasis on strategies used to couple components and interface of solvers to CAD systems. A demonstrator for power laser diodes simulation will be built, based on the coupling of three codes (both internal and commercial). A CORBA compliant approach has been selected to satisfy the needs for interoperability, platform independence and remote computing capabilities. It is expected that this study will provide valuable recommendations on interoperability issues to current and future projects. Contact Point Mr Thierry Priol INRIA/IRISA Campus Universitaire de Beaulieu F - 35042 Rennes Cedex Tel : +33-299 84 72 10 Fax : +33-299 84 71 11 Email: Thierry.Priol@irisa.fr Http : Participants SIMULOG INRIA THOMSON Country F F F Start Date 8 December 1997 Duration 18 months Octobre 1998 High Performance Computing and Networking / 201 501034 Cluster Computing for Data-Intensive Applications Keywords : High Performance Computing and Networking Technical : Information Management & Decision Support Uses : All Industrial Applications : All The aim of the tender is to stimulate industry to effectively exploit cluster computing for their data intensive applications. Cluster computing is a scalable and cost effective computation technique that integrates and optimises the use of data and processing resources in a networked environment. Data intensive applications perform intensive analysis of high-volume data sets. Work is needed 1. to consolidate lessons learnt from past and on-going experiments in the set-up and use of networked environments for high performance computing (HPC) applications. 2. to build a small reference configuration that combines data and processing resources and that provides tools to support a target data intensive application 3. to experiment with the configuration through the set-up and support of a user application 4. to disseminate widely the results of the above work through the organisation of workshops Contact Point Pietro Rossi ENEA Via Martiri di Monte Sole 4 I-40100 Bologna, Italy Participants ENEA Start Date January 1998 Octobre 1998 tel: +39 51 609 87 88 fax: +39 51 609 86 23 email: rossi@bologna.enea.it Country I Role C Duration 18 months High Performance Computing and Networking / 202 501035 DAHLIA Demonstration and assessment of HPCN in Healthcare Keywords : High Performance Computing and Networking Technical : DICOM, multimedia conferencing, ATM, CT, MRI Uses : remote image processing, remote consultation and diagnosis Industrial Applications : medical imaging applications, telemedicine New functionalities and services for healthcare applications are enabled by HPCN technologies. Powerful applications running on high performance computing facilities may be linked with high bandwidth networks to end user sites, enabling the cost of new services to be shared between several participating organizations. For example 3D/4D image reconstruction and visualisation, telediagnostics, virtual reality for simulation and training. These possibilities could be exploited to significantly improve the quality of medical services available for the citizen. To foster the take-up of HPCN technologies in healthcare applications a pilot service is set-up supporting activities of several co-operating hospitals interconnected through fast networks. The results of the experience of a 12 month pilot service will be disseminated through a web site and presentations. The pilot service selected is a multi-hospital model for Radiation Oncolgy treatment planning. One hospital acts as reference centre, and other three hospitals act as peripheral centres. Interconnection is through the Tuscany MAN at 34 Mbps. Tools are provided for management of medical data, collaborative work and multimedia conferencing, data transfer between specialists and subsidiary hospitals, communication chains within virtual teams of professionals. Contact Point Dr Giampaolo Biti Dip. Fisiopatologia Clinica–Univ of Firenze Viale Morgagni 85 I-5034 Firenze Tel : +39 055 4277264 Fax : +39 055 4379930 Email: Biti@dfc.unifi.it Http : Participants Dip. Fisiopatologia Clinica–Univ of Firenze Consorzio Pisa Ricerche AET University Hospital of Siena University Hospital of Pisa University Hospital of Arezzo Country I I I I I I Start Date January 1998 Duration 24 months Role C Sub-contractor Sub-contractor Sub-contractor Sub-contractor Sub-contractor This page is located at www.cordis.lu/esprit/src/501035.htm It was last updated on 29 October 1998, and is maintained by Massimo.Luciolli@cec.be Octobre 1998 High Performance Computing and Networking / 203 501851 SIGEX SIGnature EXploitation Keywords : High Performance Computing and Networking, Humanitarian Demining Technical : Standards, brokerage of data, Uses : Training and Validation of mine detection equipment Industrial Applications : Humanitarian Demining Context R&D teams working on multi-sensor data fusion critically need large quantities of data recorded with registered multiple sensors under realistic conditions. The gathering of such data is timeconsuming and costly. Initial signature databases will be collected by the project awarded to the EC's Joint Research Centre (JRC) as a result of the call for tender III/97/31. Other data sets have been collected in national projects or are likely to be collected in the context of running and future ESPRIT projects. The aim of this lot is to foster and facilitate exchange of such data recordings under appropriate conditions. Objectives • Define a framework to promote the exchange of APL signature measurement databases for use in selected R&D for humanitarian APL detection and identification systems. Particular emphasis is set on registered multi-sensor signature databases. • Operate a service to support this exchange for a period of 18 months. Tasks 1. Propose and reach wide agreement on a set of variables to be recorded when collecting measures and environmental conditions of APL signatures. This should include environment characteristics, types of mines and clutter, placing of mines, weather, physical parameters to be measured, range of measurements and accuracy. On-going work in this direction, among other those in the project mentioned earlier must be taken into account. Agree on a format or a set of formats to store and transmit such data. 2. Propose and reach consensus on a number of standard agreements for the provision or exchange of APL signature data to R&D teams working in humanitarian demining R&D in the context of ESPRIT and national programmes. 3. Establish a catalogue of existing landmine signature databases and of the ongoing projects developing such data, with the conditions under which the data may be made available. 4. Define procedures to validate data collected by various sources. Define methods and produce software as needed to convert data to commonly agreed formats, and to transfer them to appropriate media or through appropriate communication channels. 5. Operate a validation, distribution and brokerage service for landmine signature data to selected R&D projects of humanitarian demining for a period of 18 months. All tasks requiring agreement should aim at consulting a wide range of organisations involved in related work. In particular, these should include all related projects funded by the European Commission within ESPRIT and within the activities for scientific and technical support to Community policies. Links with non-EU projects including those supported by the United Nations are desirable. Contact Point Octobre 1998 High Performance Computing and Networking / 204 Dr. JOHN DEAN SPACE APPLICATION INSTITUTE CCR I – 21020 ISPRA Tel : +39 0 332 78 9407 Fax : +39 0 332 78 5469 Email: John.Dean@jrc.it Http Participants EC Joint Research Center Country I Start Date January 1999 Duration 24 months Octobre 1998 Role C High Performance Computing and Networking / 205 501852 MIMEVA Mine Imitations for in-field EVAluation Keywords : High Performance Computing and Networking, Humanitarian Demining Technical : Electromagnetic Signatures, Landmine surrogates Uses : Test and validation of landmine detection equipment. Industrial Applications : Humanitarian Demining Objectives Study and compare different ways to simulate landmines for research, training and development needs in humanitarian demining Tasks 1. Determine and agree with the EC, the types of mines for which surrogates are needed for research and development of equipment for humanitarian demining operations in regions of interest. 2. In the context of R&D into landmine detectors, evaluate the benefits and limitations of the different methods for replicating landmines. Define methods to validate candidate surrogate landmines. Propose and reach agreement with actors in the field on a choice. Contact Point Dr. JOHN DEAN SPACE APPLICATION INSTITUTE CCR I – 21020 ISPRA Tel : +39 0 332 78 9407 Fax : +39 0 332 78 5469 Email: John.Dean@jrc.it Http Participants EC Joint Research Center Country I Start Date January 1999 Duration 6 months Octobre 1998 Role C High Performance Computing and Networking / 206 501853 EUDEM Study on the State of the Art in the EU related to humanitarian DEMining technology, products and practice Keywords : High Performance Computing and Networking, Humanitarian Demining Technical : Survey, Demining Technology Uses : R&D in Humanitarian Demining Industrial Applications : Humanitarian Demining Objectives Study the current supply and practice in humanitarian demining. Tasks 1. Establish a list of organisations to be consulted, primarily industrial companies producing or developing equipment used in humanitarian demining actions and organisations performing or supervising humanitarian demining operations. Also include key research centres and university laboratories active in this field. 2. Perform a survey of the organisations identified as part of Task 1. The survey should exploit a combination of literature review, telephone contacts, questionnaires, face-to-face interviews, and other methods. It should start with the proposal and agreement of a plan and schedule for the work. Contact Point Prof. Jan Cornelis ETRO/IRIS Pleinlaan 2 B – 1050 BELGIUM Tel : +32 2 629 29 30 Fax : +32 2 629 28 8 3 Email: jpcornel@etro.vub.ac.be Http : Participants VUB EPFL Country B CH Start Date January 1999 Duration 6 months Octobre 1998 Role C S High Performance Computing and Networking / 207 501854 SEARCH2 SEARCH for new technologies for mine SEARCH Keywords : High Performance Computing and Networking, Humanitarian Demining Technical : Laboratory Testing, Field Testing Uses : Assessment of Technologies for Humanitarian Demining Industrial Applications : Humanitarian Demining Context A number of sensors and systems offer a potential to be applied to the detection and identification of antipersonnel landmines, even though they are not being used today in the context of humanitarian demining. They require to be tested before their potential usefulness can be evaluated and confirmed. Developers of such sensors may not have access to appropriate test facilities and are thus unable to give adequate qualification of performance in the context of humanitarian demining. In order to foster the development and take-up of new sensing techniques in humanitarian demining, tests of developed equipment or of proof-of-concept prototypes should be performed under standardised, representative conditions and in a second step under actual field conditions. Objectives • Invite candidate organisations that own sensors or systems with a potential to improve humanitarian demining performance to participate in test campaigns. Each candidate associate with an end-user in humanitarian demining to develop a common understanding of the validity of tests performed. • Select a set of sensors and test campaigns to cover at least 4 sensing modalities not being used today in routine humanitarian demining. The sensors or prototypes must have a defined route to exploitation showing improvements in cost, speed and safety with respect to current practice. The campaigns must all cover first-phase tests in controlled conditions. Two of the sensors must also be subjected to second-phase tests in real field conditions. • Contract the test activities with the selected owners of sensor systems or prototypes and, as the case may be, support the test activities. • Consolidate and disseminate the results to reach a majority of main EU actors and donors in the field of humanitarian demining. Tasks 1. Organise access and/or provide access to test facilities for verification, validation and benchmarking of systems and equipment for the detection and identification of APL. The access should enable 4 test periods of 1 week each in controlled conditions for the first phase tests and 2 test periods of 2 weeks each in field conditions. Bidders will have access to facilities already contracted in Tender III/97/31 Lot 1, awarded to the Commission's Joint Research Centre. 2. Organise a call for test participants. Evaluate the proposals, propose a list ordered by priorities and agree with the Commission, a selection of the most suitable candidates. 3. Contract and manage the agreed test campaigns, including possible provision of assistance to organisations for their support during the conduct of the campaigns. 4. Compile the results of the tests and disseminate these to the intended audience. Contact Point Octobre 1998 High Performance Computing and Networking / 208 Dr. JOHN DEAN SPACE APPLICATION INSTITUTE CCR I – 21020 ISPRA Tel : +39 0 332 78 9407 Fax : +39 0 332 78 5469 Email: John.Dean@jrc.it Http Participants EC Joint Research Center Country I Start Date January 1999 Duration 18 months Octobre 1998 Role C High Performance Computing and Networking / 209 977069 EMG-NET Research Network for intelligent support of EMG studies Keywords : High Performance Computing and Networking Technical : machine learning, data mining, distributed databases, web service Uses : diagnostics Industrial Applications : Health services The project will establish a research network combining IT and medical teams across Europe, dedicated to improve quality and efficiency of medical practice in Electromyography (EMG). This project aims at progressing activities started in the ESTEEM project of the AIM programme, and in extending them to participants from the central European countries. The work carried out in the network is aimed at: – Sharing both the acquired EMG expertise and the previously accumulated data; – Extracting a useful knowledge encoded into the available EMG data; – Achieving a complete standardisation of the knowledge available in the medical nodes; – Disseminating the current achievements in EMG related and IT studies. Use of web technology is foreseen to enable an easy and user-friendly access to information and tools. Contact Point Dr Danielle Ziebelin SHERPA – INRIA Rhône-Alpes 655 Avenue de l’Europe F-38330 Montbonnot Saint-Martin Tel : +33 4 76 61 53 68 Fax : +33 4 76 51 52 07 Email: Danielle.Ziebelin@inrialpes.fr http : Participants INRIA Rhône-Alpes IT Institute-Bulgarian Academy of Sciences Gentofte Hospital Unione Italiana Lotta alla Distrofia Muscolare Spitalul Clinic N GH Lupu Electrical Engineering Faculty - LIMBISS Universidade de Lisboa University of Skopje Johannes Gutenburg Univ – Mainz Comenius University – Medical School RNC – Newcastle General Hospital AI Research Lab First City Hospital Univ Joseph Fourier Slovak University of technology Country F BG DK I Role C P P Sub-contractor RO MNI P MNI D SK UK RO BG F SK Sub-contractor Sub-contractor Sub-contractor Sub-contractor Sub-contractor Sub-contractor Sub-contractor Sub-contractor Sub-contractor Sub-contractor Sub-contractor Start Date 1 October 1998 Duration 24 months This page is located at www.cordis.lu/esprit/src/977069.htm It was last updated on 29 October 1998, and is maintained by Massimo.Luciolli@cec.be Octobre 1998 High Performance Computing and Networking / 210 INCO 977113 KNIXMAS Knowledge Shared XPS-Based Research Network Using Multi-Agent-Systems Keywords: high performance computing, intelligent research network, decision support system, intelligent self adapting knowledge retrieval, multi agent systems, decision supported collaborative working Technical: AI-modules, knowledge broker modules, decision support modules, Corba standards Application: distributed management systems, collaborative research and development system, enterprise engineering Due to the increasing demand of globally distributed information within complex researching and developing there is a need for intelligent self adapting global research and development networks, especially under GroupWare aspects. The main objective is to design and establish a framework (shell) for an intelligent global research network consisting of connected local (national) working and research networks. This framework will consist of distributed co-operating expert systems (Intelligent Base Systems) with decision support modules for research management and intelligent knowledge retrieval provided by applied multi-agent systems. For the prototype the local research networks and working spaces will be realized as Intranets in 5 different European countries (Germany, Finland, Hungary, Poland, Bulgaria). The center of each local network will be an intelligent base system conceived as expert system with co-operating and communication facilities via Internet/Intranet. The entire knowledge will be shared over the whole network and managed by Intelligent Knowledge Brokers. The main characteristics of the research network are the following: The entire network is supplied with AI-components and methods. Modularity and flexibility will be guaranteed by structuring and designing the whole system in an object oriented way. Adaptiveness and flexible adjustment will be provided by reflecting and evaluating user response and acceptance data via complex feedback and influence loops. Developing this research network as a modular AI-system ensures a high degree of transferability and adjustment capabilities. Using dynamic knowledge and databases and digital repositories controlled by intelligent knowledge brokers guarantees inter- and exchange of specialized knowledge. For the intelligent knowledge retrieval global and local standard multiagent systems will be installed on heterogeneous platforms using CORBA standards. The flexible modular structure of the framework gives the possibility to apply the system to others than technical or scientific areas. It will be universally applicable for several areas in industry and economics. At first it is planned to process the knowledge of high-tech areas like control theory and communication techniques. In addition several testing partners are involved to test the system in other than scientific areas (customizing of SPP/(R/3) software, enterprise modeling, Tele teaching etc.). Contact Point H.-J. Nern University of Wuppertal Group of Automatic Control and Technical Cybernetics Fuhlrottstr. 10 D-42097 Wuppertal Octobre 1998 tel: +49 202 439 2950 fax: +49 202 439 2953 email: nern@uni-wuppertal.de http : to be announced High Performance Computing and Networking / 211 Participants Prof. Dr. sc.techn. H.A. Nour Eldin (Institute of Automatic Control and Technical Cybernetics, BUGHW) Prof. Dr. sci. J. D. Zaprianov (Institute of Control and Systems Research, BAS) Prof. Dr. hab. A. Dziech (Telecommunication Department, AGH) Prof. Dr. Laszlo Kovacs (MTA SZTAKI, Hungarian Academy of Science) Prof. Dr. Matti Hämäläinen (Espoo-Vantaa Institute of Technology, Espoo) Prof. Dr.-Ing. J. Wassermann (Institute of Applied Science, FHD Stuttgart) Start Date October 1998 Country D Role C BG P PL P HU P FIN P D P Duration 24 months This page is located at www.cordis.lu/esprit/src/977113.htm It was last updated on 13 October 1998, and is maintained by Merce.Griera-I-Fisa@dg3.cec.be Octobre 1998 High Performance Computing and Networking / 212 Index of Projects by Number Project# Acronym Title 20018 PINEAPL 20074 HPC-VAO 20082 20089 FRONTIER MERCURY 20111 FSI-SD 20115 20116 20118 PAROS ODESIM MEMIPS 20124 20161 ESOP CISPAR 20162 20170 20183 20184 20189 PHAROS EMCP2 HOISE NM HPS-ICE PROMENVIR 20216 INSIDE 20231 20248 20255 20259 TOOLSHED CLOVIS ISIS APEX 20956 JAMES-RN 20957 20966 21007 TEN-34 MICA IOTA 21012 21017 EFTOS INNOVA 21023 21028 21036 21037 21040 21042 21050 21073 CATIE PACHA EIVIS PACOS 2 EUROPRO CTIAC SHAROC 3DULTRA 21089 21102 21111 22503 HIPERTEX EUROPORT-D HPC-STANDARDS EROPPA 22505 WARPSPEID 22508 22509 22517 MSC ESD_NET VISTA 22519 22528 22542 22556 22559 22684 RCNET CANET COVAS APAT COLBA EPOCH PARALLEL INDUSTRIAL NUMERICAL APPLICATIONS AND PORTABLE LIBRARIES........................................................................................................................4 HIGH PERFORMANCE COMPUTATIONAL ENVIRONMENT FOR VIBRO-ACOUSTIC OPTIMIZATION ................................................................................................................5 OPEN SYSTEM FOR COLLABORTIVE DESIGN.............................................................6 PERFORMANCE MANAGEMENT OF COMMERCIAL PARALLEL DATABASE SYSTEMS ..........................................................................................................................7 FLUID STRUCTURE INTERACTION- HIGH PERFORMANCE SIMULATION FOR STRUCTURAL DESIGN....................................................................................................8 PARALLEL LARGE SCALE AUTOMATIC SCHEDULING..............................................9 OPTIMUM DESIGN OF MULTIBODY SYSTEMS..........................................................10 MODEL ENHANCED SOLUTION METHODS FOR INTEGER PROGRAMMING SOFTWARE .....................................................................................................................11 A PARALLEL OPTIMISATION SCHEME APPLIED TO EXTRUSION DIE DESIGN....12 OPEN INTERFACE FOR COUPLING OF INDUSTRIAL SIMULATION CODES ON PARALLEL SYSTEMS....................................................................................................13 OPEN HPF PROGRAMMING ENVIRONMENT..............................................................14 ELECTROMAGNETIC COMPATIBILITY USING PARALLEL PARAMETERISATION15 HPCN EUROPE ON LINE NEWSMAGAZINE................................................................16 HIGH PERFORMANCE SIMULATION OF INTERNAL COMBUSTION ENGINES.......17 HIGH PERFORMANCE COMPUTER BASED PROBALISTIC MECHANICAL DESIGN ENVIROMENT ................................................................................................................18 INTEGRATED SIMULATION AND DESIGN SYSTEM FOR CIVIL AND STRUCTURAL ENGINEERING................................................................................................................19 TOOLS FOR HIGH PRODUCTIVITY ENGINEERING DESIGN.....................................20 COMPUTING LARGE OBJECTS VISUALISATION INFORMATION SYSTEM............21 INTERACTIVE SATELLITE IMAGE SERVER ON GENERATION IMAGE CODING...22 HIGH PERFORMANCE COMPUTING FOR PROCESS INDUSTRIES AWARENESS AND PROMOTION EXERCISE.......................................................................................23 INTERCONNECTION OF EUROPEAN RESEARCH AND UNIVERSITY NETWORKS AT 34-155 MBIT/S ...........................................................................................................24 TRANS-EUROPEAN NETWORK INTERCONNECT AT 34-155 MBPS..........................26 MODEL FOR INDUSTRIAL CFD APPLICATIONS.........................................................27 IOTA-INTELLIGENT OBJECT TRACKING AND ANALYSIS USING SMART SENSOR TECHNOLOGY................................................................................................................28 EMBEDDED FAULT TOLERANT SUPERCOMPUTING................................................29 HIGH PERFORMANCE TELECONTROL STATION WITH IMAGE AND SOUND CAPABILITIES ................................................................................................................30 COLOUR AND TEXTURE INSPECTION EQUIPMENT.................................................31 PARALLEL SCALABLE COMPUTERS FOR HIGH PERFORMANCE...........................32 EMBEDDED INTERACTIVE VIDEO SERVER...............................................................33 PARALLEL COMPUTING FOR SPAIN 2........................................................................34 EUROPEAN PROCESSING CORE TECHNOLOGY (EUROPRO)...................................45 COMPUTATIONAL TOOLS AND INDUSTRIAL APPLICATIONS OF COMPLEXITY.46 SYSTEM FOR HIGH-SPEED FOR AUTOMATIC READING OF CHEQUES..................47 HIGH-PERFORMANCE COMPUTING FOR CLINICAL APPLICATIONS OF 3D ULTRASOUND................................................................................................................48 HIGH PERFORMANCE SYSTEM FOR TEXTILE EVALUATION..................................49 EUROPORT DISSEMINATION AND DEMONSTRATION.............................................50 HPC-STANDARDS ..........................................................................................................51 EXPERIMENTS ON REMOTE OPERATIONS FOR POST PRODUCTION APPLICATIONS...............................................................................................................52 WIDE AREA PROJECT SUPPORTING PAN-EUROPEAN INFORMATION DISSEMINATION............................................................................................................53 MULTI-SITE COMPUTING.............................................................................................55 HIGH PERFOMANCE ELECTRONIC SYSTEMS DESIGN NETWORK.........................56 VIRTUAL INTERACTIVE STUDIO TELEVISION APPLICATION USING NETWORKED GRAPHICAL SUPERCOMPUTERS........................................................57 EXPLOITING HPCN IN AN ENGINEERING CONSULTANCY ENVIRONMENT.........58 COLLABORATIVE AUTOMOTIVE NETWORK............................................................59 COLLABORATIVE VISUALIZATION OF AUTOMOTIVE SIMULATIONS..................60 APPLICATIONS PROFESSIONELLES SUR ATM..........................................................61 CUSTOMER ONLINE BEHAVIOUR ANALYSIS............................................................62 ELECTROMAGNETIC PRODUCT OPTIMISATION USING COMPUTERS WITH HIGH PERFORMANCE..............................................................................................................63 Octobre 1998 Page High Performance Computing and Networking / 213 22685 MESODYN 22689 SIMTRAP 22691 SEEDS 22693 22695 HYPERBANK SLOEGAT 22696 22700 HAT CRITIKAL 22707 22709 22716 22717 HP-RETAIL FAST-SIM DORO OPTIMUM 22723 HITERM 22727 22740 WEPTEL PARMAT 22745 22746 INDEX MEDUSA 22752 ACTIVATE 22765 22835 23174 DESIRE II NETAPDEX SISCI 23475 RADAR 23480 FLEXSECURE 23483 23486 23499 23502 23516 23713 23770 24003 24396 24404 24406 24410 24414 EPSIS PHASE DYNAMITE FITS WINPAR TTN-T FINNOVA HIPERTTN ATTN ENTICE METIER2 METIER1 NOT SO MAD 24428 24448 TTNMV-SF CAPRICE 24462 24484 24617 24620 24621 24722 24724 24727 24740 24742 24757 24758 TTNATGMD INNO DANHIT TETRAPC ICETACT BIOTITAN PROHPC PDC TTN AVS-TTN DUTCH TTN CEPBA-TTN ESCALATE 24897 HCHLOUSO 24900 24903 HPCN-SCAT FLASH Octobre 1998 SOFTWARE FOR MESOSCALE DYNAMICS SIMULATIONS ON HPCN PLATFORMS IN CHEMICAL ENGINEERING.......................................................................................64 HPCN SIMULATION OF TRAFFIC INDUCED AIR POLLUTION USING PARALLEL COMPUTING IN A DISTRIBUTED NETWORK.............................................................65 SIMULATION ENVIRONMENT FOR THE EVALUATION OF DISTRIBUTED TRAFFIC CONTROL SYSTEMS......................................................................................................66 HIGH PERFORMANCE BANKING.................................................................................67 SHORT AND LONG TERM OPTIMIZATION OF ELECTRICITY GENERATION AND TRADING ........................................................................................................................68 HIGH PERFORMANCE ARBITRAGE DETECTION&TRADING...................................69 CLIENT-SERVER RULE INDUCTION TECHNOLOGY FOR INDUSTRIAL KNOWLEDGE ACQUISITION FROM LARGE DATABASES........................................70 HIGH PERFORMANCE PARALLEL COMPUTING FOR MEDIUM SIZED RETAILERS71 MMP SYSTEM FOR SIMULATED NMR DATA ANALYSIS..........................................72 DOCUMENT ROUTING..................................................................................................73 APPLYING HPC TECHNIQUES TO THE INDUSTRIAL DESIGN : AUTOMATIC OPTIMIZATION OF FORGING PROCESSES.................................................................74 HIGH-PERFORMANCE COMPUTING AND NETWORKING FOR TECHNOLOGICAL AND ENVIRONMENTAL RISK MANAGEMENT ..........................................................75 INNOVATIVE WEATHER PRESENTATION ON TELEVISION.....................................76 EFFICIENT HANDLING OF LARGE MATRICES ON HIGH PARALLEL COMPUTER SYSTEMS WITH THE PERMAS CODE..........................................................................77 INTELLIGENT DATA EXTRACTION.............................................................................78 MULTI-DISCIPLINARY ENGINEERING DESIGN VIA UNITARY SOFTWARE APPLICATIONS...............................................................................................................79 ADVANCED COMPUTING TECHNOLOGY FOR INNOVATIVE VIBRO-ACOUSTIC TOOLS IN ENGINEERING..............................................................................................80 DESIGN BY SIMULATION AND RENDERING ON PARALLEL ARCHITECTURES II81 NETWORK APPLICATION AND DISSEMINATION EXPERTISE.................................82 STANDARD SOFTWARE INFRASTRUCTURE FOR SCI-BASED PARALLEL SYSTEMS ........................................................................................................................83 HIGH PERFORMANCE RAINFALL RADAR IMAGE PROCESSING FOR SEWAGE SYSTEMS CONTROL......................................................................................................84 MODULAR, SCALABLE DETECTION SYSTEM FOR EXPLOSIVES, DRUGS AND WEAPONS IN HAND BAGGAGE AND MEDIUM SIZED PARCELS.............................85 REAL-TIME ANALYSIS OF VIDEO IMAGE STRUCTURE...........................................86 A DISTRIBUTED PHARMACEUTICAL APPLICATIONS SERVER...............................87 DYNAMIC TASK MIGRATION EXECUTION ENVIRONMENT...................................88 FORTRAN INTEGRATED TOOL SET.............................................................................89 WINDOWS BASED PARALLEL COMPUTING..............................................................90 TECHNOLOGY TRANSFER NODE THURINGIA...........................................................91 NOVEL APPROACHES TO FINNISH INDUSTRIAL COMPUTING...............................92 HIPERCOSME TTN.........................................................................................................93 AUSTRO-TTN..................................................................................................................94 EUROPEAN NEW TECHNOLOGY FOR INDUSTRIAL & COMMERCIAL END-USERS95 MECHANISM FOR ENABLING HPCN TECHNOLOGY TRANSFER IN EUROPE 2.....96 MECHANISM FOR ENABLING HPCN TECHNOLOGY TRANSFER IN EUROPE 1.....97 NODO DI TRANSFERIMIENTO TECNOLOGICO A SISTEMI DI OPERATORI DEL MANIFATTURIERO E DEI SERVIZI..............................................................................98 TECHNOLOGY TRANSFER NODE OF MACHINE VISION IN FINLAND....................99 CENTER FOR ACCESSING HPCN BEST PRACTICE INFORMATION, TECHNOLOGY AND APPLICATION ENVIRONMENTS.......................................................................100 TECHNOLOGY TRANSFER NODE AT GMD .............................................................. 101 AN INSPECTION TECHNOLOGY TRANSFER NODE.................................................102 DANISH HPCN TECHNOLOGY TRANSFER NODE.................................................... 103 TECHNOLOGY TRANSFER IN PARALLEL COMPUTING.........................................104 IRISH CENTRE FOR PROACTIVE TRANSFER OF ADVANCED COMPUTING.........105 TRANSFER OF BIOINFORMATICS INTO EUROPEAN INDUSTRY........................... 106 PROMOTION OF HIGH PERFORMANCE COMPUTING.............................................107 SWEDISH HPCN TECHNOLOGY TRANSFER NODE AT PDC...................................108 ADVANCED VIDEOSURVEILLANCE-TECHNOLOGY TRANSFER NODE...............109 DUTCH HPCN TRANSFER TECHNOLOGY NODE..................................................... 110 CEPBA-TTN................................................................................................................... 111 EXPLOITATION OF SYNERGIES AND CO-OPERATIVE ACTIONS LEADING TO ACCELERATED TAKE-UP AND EFFECTIVENESS .................................................... 112 HYDROCARBON AND CHEMICAL LOGISTICS OPTIMIZATION UNDER UNCERTAINTY VIA STOCHASTIC OPTIMIZATION.................................................113 HPCN STOCHASTIC CORRELATION OF ANALYSIS AND TEST.............................. 114 HPCN TOOLS FOR ENHANCED HYDRODYNAMIC DESIGN OF FAST SHIPS ON PARALLEL COMPUTING PLATFORMS......................................................................115 High Performance Computing and Networking / 214 24907 24916 24949 24950 24959 24960 24977 24986 25000 25009 25047 25050 25058 25059 26255 26267 26276 26285 26286 26287 26293 26313 26317 26329 26331 26337 26342 26347 26356 26386 26387 26700 26749 27021 27741 28144 28159 28168 28189 28195 28245 28247 28249 28283 28350 28363 28372 28410 28425 28842 Octobre 1998 COLOMBO PARALLEL COMPUTERS IMPROVE CLEAN UP OF SOILS BY MODELLING BIORMEDIATION.........................................................................................................116 VICAR VIDEO INDEXING CLASSIFICATION ANNOTATION AND RETRIEVAL.................117 IMPACT INVERSE METHODS FOR WAVE PROPAGATION APPLICATIONS IN TIME DOMAIN118 OPERA OPERATORS TRAINING DISTRIBUTED REAL-TIME SIMULATIONS..................... 120 CRIPS-DM CROSS INDUSTRY STANDARD PROCESS FOR DATA MINING............................... 121 PARROT PARALLEL CREW ROSTERING ..................................................................................122 STRETCH STORAGE AND RETRIEVAL BY CONTENT OF IMAGED DOCUMENTS.................123 DEDEMAS DECENTRALISED DECISION MAKING AND SCHEDULING....................................124 HIPSID HIGH PERFORMANCE SIMULATION FOR INTERACTIVE DESIGN........................ 125 DOMINOS DOMAIN DECOMPOSITION METHODS FOR INTEGRATED NOISE AND STRUCTURAL ANALYSIS........................................................................................... 127 QUASI QUANTUM SIMULATION IN INDUSTRY...................................................................128 JULIUS JOINT INDUSTRIAL INTERFACE FOR END-USERS SIMULATIONS........................ 129 DECISION HPCN INTEGRATED OPTIMIZATION STRATEGIES FOR INCREASED ENGINEERING DESIGN COMPLEXITY......................................................................130 CHAMAD COMPLEXE HEXAHEDRAL AND ADAPTATIVE MESHING OF ARBITRARY DOMAINS...................................................................................................................... 131 GABBY ANIMATOR GABBY ANIMATOR : HOW TO RISE THE PRODUCTIVITY IN THE ANIMATION MARKET ....................................................................................................................... 132 SCHUMANN SUPPLY CHAIN UNCERTAINTY MANAGEMENT NETWORK OPTIMIZATION .....133 SEP-TOOLS AN INTER-OPERABLE SUITE OF EUROPEAN HPCN TOOLS...................................134 CAVALCADE COLLABORATIVE VIRTUAL CONSTRUCTION AND DESIGN.................................136 VINE VIDEO ON DEMAND AND INTERACTIVE VIDEO EDITING SYSTEM..................... 137 M3D MULTI-SITE COOPERATIVE 3D DESIGN SYSTEM FOR ARCHITECTURE..............138 MINEREC DEMONSTRATION OF REAL TIME MINE RECOGNITION PROCEDURES ANALYSING GPR DATA ON EUROPRO.....................................................................139 CIPRESS COMPLEX INFORMATION PATTERNS RETRIEVAL WITH A PARALLEL DISTRIBUTED PROCESSING KNOWLEDGE ENGINE SEARCH SYSTEM................140 EUROSTORE A HIGH PERFORMANCE STORAGE PROJECT........................................................... 141 MISSIMU MINIMALLY INVASIVE SURGICAL SIMULATOR.................................................... 142 DREAM DATA FUSION AS A REMEDY AGAINST MINES ...................................................... 143 GEODE GROUND EXPLOSIVE ORDNANCE DETECTION SYSTEM-MULTISENSOR FUSION SYSTEM FOR APL DETECTION LOCALIZATION AND CLASSIFICATION.............144 NOVICE NETWORK ORIENTED VISUALIZATION IN A CLINICAL ENVIRONMENT............145 EDISON EUROPEAN ADVANCED DISTRIBUTION SIMULATION ENVIRONMENT.............146 POEM PARALLEL PROCESSING OF VOLUMINOUS EDIFACT DOCUMENTS................... 147 DISTAL DISTRIBUTED SOFTWARE ACCESS FOR LARGE SCALE ENGINEERING APPLICATIONS.............................................................................................................148 DYNALOG DYNAMIC LOGISTICS FOR DISTRIBUTION NETWORKS........................................149 ELTRAMOS ELECTRONIC MARKET AND TRADING OF MODELLING SERVICES AND ASSETS OF ENGINEERING SMES AND INSTITUTES.............................................................. 150 MIMIC MINING THE INTERNET FOR MARKETING INTELLIGENCE..................................151 COSMOS CONSTRUCTION SITE MOBILE OPERATIONS SUPPORT........................................152 EUROTOOLS WORKING GROUP FOR EUROPEAN HPCN TOOLS PROMOTION........................... 154 DECAST ENHANCED DESIGN ENVIRONEMENT FOR INDUSTRIAL CASTING PROCESSES ON PARALLEL COMPUTING PLATFORMS ............................................................... 155 ASRA-HPC ADVANCED STRUCTURAL RELIABILITY ANALYSIS ON HIGH PERFORMANCE COMPUTERS.................................................................................................................157 ALICE QUANTITATIVE FLOW FIELD VISUALIZATION ...................................................... 158 ALESSIA APPLICATION OF LARGE EDDY SIMULATION TO THE SOLUTION OF INDUSTRIAL PROBLEMS............................................................................................ 159 AMEGOS AUTOMATIC MESH GENERATION AND OPTIMISATION FOR INDUSTRIAL FLOW SIMULATIONS..............................................................................................................160 RAMSES REGIONAL EARTH OBSERVATION APPLICATION FOR MEDITERRANEAN SEA EMERGENCY SURVEILLANCE................................................................................... 161 VISIT VIRTUAL INTUITIVE SIMULATION TESTBED......................................................... 162 HISTRIDE A HIGH PERFORMANCE STRUCTURAL IDENTIFICATION ENVIRONEMENT......163 HEATMASTER INDUCTION HEATING MASTERING FOR METAL PROCESSING THROUGH INVERSE ANALYSIS....................................................................................................164 BLOODSIM HPCN-ENABLED SIMULATION OF CARDIOVASCULAR AND OTHER BIOMEDICAL FLOW............................................................................................................................. 165 MADS MULTI-PURPOSE ANTENNA DESIGN SIMULATOR.................................................167 HPNURSA HIGH PERFORMANCE NUMERICAL UNSTEADY ROTOR STATOR ANALYSIS....168 ROXY THE INTERNET LIVE INITIATIVE..............................................................................169 BISANTE BROADBAND INTEGRATED SATELLITE NETWORK TRAFFIC EVALUATION............................................................................... 170 KLEE&CO KNOWLEDGE AND LEARNINGS ENVIRONMENTS FOR EUROPEAN AND CREATIVE ORGANIZATIONS ..................................................................................... 172 High Performance Computing and Networking / 215 28928 28942 29255 29381 KNOWNET MOST KLIF ARIS 29479 MACADAM 29732 29737 29812 CONTENTS HPGIN LOTUS 29870 29895 HOPE PICE 29902 DEMINE 29907 NEPHEW 29909 METODIS 29944 INFIELD 29958 MINESEYE 85386 85387 PROMISE RAIN 85388 85389 85390 85391 85392 VR FOR EUROPE 501031 MINETEST 501032 501033 MINESIGN SOFT-IT 501034 501035 501851 501852 501853 DAHLIA SIGEX MIMEVA EUDEM 501854 977069 977113 SEARCH² EMG-NET KNIXMAS Octobre 1998 KNOWLEDGE MANAGEMENT WITH INTRANET TECHNOLOGIES ....................... 173 MEDICAL REPRESENTATIVE ON-LINE SIMULATION TRAINING.......................... 175 A KNOWLEDGE AND LEARNING INFRASTRUCTURE.............................................177 ACTION FOR RESEARCH AND INFORMATION SUPPORT IN CIVILIAN DEMINING A NETWORK OF EXCELLENCE TO HELP IMPROVE THE EFFECTIVENESS OF R&D ACTIVITIES IN DEMINING TECHNOL........................................................................178 MULTISENSOR ACQUISITION CAMPAIGN FOR ANALYSIS AND DATA FUSION OF ANTIPERSONAL MINESS............................................................................................ 179 COMPONENTS TECHNOLOGY EXERCISED ON NT SYSTEMS................................ 180 HIGH PERFORMANCE GIGABIT I2O NETWORKING SOFTWARE........................... 181 LIGHT ORDNANCE DETECTION BY TELEOPERATED UNMANNED SYSTEM................................................................. 182 HAND-HELD OPERATIONAL DE-MINING SYSTEM............... 183 PINPOINT, IDENTIFICATION, CLEARANCE AND ENSURANCE185 IMPROVED COST-EFFICIENT SURFACE PENETRATING RADAR DETECTOR WITH SYSTEM ON CHIP SOLUTION FOR HUMANITARIAN DEMINING (DEMINE).................................. 186 NETWORK OF PC'S HETEROGENEOUS WINDOWS-NT ENGINEERING TOOLSET ........................................................... 187 METACOMPUTING TOOLS FOR DISTRIBUTED SYSTEMS.... 189 INDUSTRY NGO FIELDTEST OF IMPROVED EFFICIENCY MULTISENSOR LANDMINE DETECTOR.................................. 190 DEVELOPMENT AND OPTIMIZATION OF A DUAL SENSOR SYSTEM WITH REAL TIME DIGITAL SIGNAL PROCESSING FOR THE DETECTION AND IDENTIFICATION OF BURIED LANDMINES AND UNEXP.......................................................... 191 DEMONSTRATION AND ASSESSMENT OF HPCN IN MEDICAL IMAGING............192 DEMONSTRATION AND ASSESSMENT OF HPCN IN NEURAL NETWORK APPLICATIONS FOR INDUSTRY AND MEDICINE.................................................... 193 DEMONSTRATION AND ASSESSMENT OF HPCN TOOLS FOR MESH GENERATION AND MESH PARTITIONING ........................................................................................ 194 DEMONSTRATION AND ASSESSMENT OF SCALABLE LOW-COST HPCN SYSTEMS FOR SIMULATION APPLICATIONS IN NON-TRADITIONAL INDUSTRIAL SECTORS195 DEMONSTRATION AND ASSESSMENT OF HPCN FOR INFORMATION MANAGEMENT AND DECISION SUPPORT APPLICATIONS....................................196 .......................................................................................................................................197 DEMONSTRATION AND ASSESSMENT OF HPCN IN MULTI-SITE APPLICATIONS FOR VIRTUAL REALITY.............................................................................................. 198 SUPPORT ACTIVITIES TO ESPRIT R&D PROJECTS RELATING TO HUMANITARIAN DEMINING .................................................................................................................... 199 MEASUREMENT CAMPAIGN IN SUPPORT OF HUMANITARIAN DEMINING.......200 SOFTWARE INTEROPERABILITY AND PLATFORM INDEPENDENCE: THE NEXT GENERATION OF SIMULATION ENVIRONMENTS................................................... 201 CLUSTER COMPUTING FOR DATA-INTENSIVE APPLICATIONS........................... 202 DEMONSTRATION AND ASSESSMENT OF HPCN IN HEALTHCARE...................... 203 SIGNATURE EXPLOITATION...................................................................................... 204 MINE IMITATIONS FOR IN-FIELD EVALUATION..................................................... 206 STUDY ON THE STATE OF THE ART IN THE EU RELATED TO HUMANITARIAN DEMINING TECHNOLOGY, PRODUCTS AND PRACTICE........................................207 SEARCH FOR NEW TECHNOLOGIES FOR MINE SEARCH.......................................208 RESEARCH NETWORK FOR INTELLIGENT SUPPORT OF EMG STUDIES..............210 KNOWLEDGE SHARED XPS-BASED RESEARCH NETWORK USING MULTI-AGENT SYSTEMS ...................................................................................................................... 211 High Performance Computing and Networking / 216 Index of Projects by Acronym Acronym Project # Title Page 21073 3DULTRA HIGH-PERFORMANCE COMPUTING FOR CLINICAL APPLICATIONS OF 3D ULTRASOUND................................................................................................................48 22752 ACTIVATE 28189 ALESSIA 28168 28195 ALICE AMEGOS 22556 20259 APAT APEX 29381 ARIS 28159 ASRA-HPC 24396 24740 ATTN AVS-TTN ADVANCED COMPUTING TECHNOLOGY FOR INNOVATIVE VIBRO-ACOUSTIC TOOLS IN ENGINEERING..............................................................................................80 APPLICATION OF LARGE EDDY SIMULATION TO THE SOLUTION OF INDUSTRIAL PROBLEMS............................................................................................ 159 QUANTITATIVE FLOW FIELD VISUALIZATION ...................................................... 158 AUTOMATIC MESH GENERATION AND OPTIMISATION FOR INDUSTRIAL FLOW SIMULATIONS..............................................................................................................160 APPLICATIONS PROFESSIONELLES SUR ATM..........................................................61 HIGH PERFORMANCE COMPUTING FOR PROCESS INDUSTRIES AWARENESS AND PROMOTION EXERCISE.......................................................................................23 ACTION FOR RESEARCH AND INFORMATION SUPPORT IN CIVILIAN DEMINING A NETWORK OF EXCELLENCE TO HELP IMPROVE THE EFFECTIVENESS OF R&D ACTIVITIES IN DEMINING TECHNOL........................................................................178 ADVANCED STRUCTURAL RELIABILITY ANALYSIS ON HIGH PERFORMANCE COMPUTERS.................................................................................................................157 AUSTRO-TTN..................................................................................................................94 ADVANCED VIDEOSURVEILLANCE-TECHNOLOGY TRANSFER NODE...............109 24722 28425 BIOTITAN BISANTE TRANSFER OF BIOINFORMATICS INTO EUROPEAN INDUSTRY........................... 106 28350 BLOODSIM HPCN-ENABLED SIMULATION OF CARDIOVASCULAR AND OTHER BIOMEDICAL FLOW............................................................................................................................. 165 22528 24448 CANET CAPRICE 21023 26285 24757 25059 CATIE CAVALCADE CEPBA-TTN CHAMAD 26313 CIPRESS 20161 CISPAR 20248 22559 24907 CLOVIS COLBA COLOMBO 29732 27021 22542 24959 22700 CONTENTS COSMOS COVAS CRIPS-DM CRITIKAL 21042 CTIAC COLLABORATIVE AUTOMOTIVE NETWORK............................................................59 CENTER FOR ACCESSING HPCN BEST PRACTICE INFORMATION, TECHNOLOGY AND APPLICATION ENVIRONMENTS.......................................................................100 COLOUR AND TEXTURE INSPECTION EQUIPMENT.................................................31 COLLABORATIVE VIRTUAL CONSTRUCTION AND DESIGN.................................136 CEPBA-TTN................................................................................................................... 111 COMPLEXE HEXAHEDRAL AND ADAPTATIVE MESHING OF ARBITRARY DOMAINS...................................................................................................................... 131 COMPLEX INFORMATION PATTERNS RETRIEVAL WITH A PARALLEL DISTRIBUTED PROCESSING KNOWLEDGE ENGINE SEARCH SYSTEM................140 OPEN INTERFACE FOR COUPLING OF INDUSTRIAL SIMULATION CODES ON PARALLEL SYSTEMS....................................................................................................13 COMPUTING LARGE OBJECTS VISUALISATION INFORMATION SYSTEM............21 CUSTOMER ONLINE BEHAVIOUR ANALYSIS............................................................62 PARALLEL COMPUTERS IMPROVE CLEAN UP OF SOILS BY MODELLING BIORMEDIATION.........................................................................................................116 COMPONENTS TECHNOLOGY EXERCISED ON NT SYSTEMS................................ 180 CONSTRUCTION SITE MOBILE OPERATIONS SUPPORT........................................152 COLLABORATIVE VISUALIZATION OF AUTOMOTIVE SIMULATIONS..................60 CROSS INDUSTRY STANDARD PROCESS FOR DATA MINING............................... 121 CLIENT-SERVER RULE INDUCTION TECHNOLOGY FORINDUSTRIAL KNOWLEDGE ACQUISITION FROM LARGE DATABASES........................................70 COMPUTATIONAL TOOLS AND INDUSTRIAL APPLICATIONS OF COMPLEXITY.46 501035 24617 28144 DAHLIA DANHIT DECAST 25058 DECISION 24986 DEDEMAS -A- -BBROADBAND INTEGRATED SATELLITE NETWORK TRAFFIC EVALUATION............................................................................... 170 -C- -D- Octobre 1998 DEMONSTRATION AND ASSESSMENT OF HPCN IN HEALTHCARE...................... 203 DANISH HPCN TECHNOLOGY TRANSFER NODE.................................................... 103 ENHANCED DESIGN ENVIRONEMENT FOR INDUSTRIAL CASTING PROCESSES ON PARALLEL COMPUTING PLATFORMS ............................................................... 155 HPCN INTEGRATED OPTIMIZATION STRATEGIES FOR INCREASED ENGINEERING DESIGN COMPLEXITY......................................................................130 DECENTRALISED DECISION MAKING AND SCHEDULING....................................124 High Performance Computing and Networking / 217 29902 DEMINE IMPROVED COST-EFFICIENT SURFACE PENETRATING RADAR DETECTOR WITH SYSTEM ON CHIP SOLUTION FOR HUMANITARIAN DEMINING (DEMINE).................................. 186 22765 26386 DESIRE II DISTAL 25009 DOMINOS 22716 26331 24742 26387 23499 DORO DREAM DUTCH TTN DYNALOG DYNAMITE DESIGN BY SIMULATION AND RENDERING ON PARALLEL ARCHITECTURES II81 DISTRIBUTED SOFTWARE ACCESS FOR LARGE SCALE ENGINEERING APPLICATIONS.............................................................................................................148 DOMAIN DECOMPOSITION METHODS FOR INTEGRATED NOISE AND STRUCTURAL ANALYSIS........................................................................................... 127 DOCUMENT ROUTING..................................................................................................73 DATA FUSION AS A REMEDY AGAINST MINES ...................................................... 143 DUTCH HPCN TRANSFER TECHNOLOGY NODE..................................................... 110 DYNAMIC LOGISTICS FOR DISTRIBUTION NETWORKS........................................149 DYNAMIC TASK MIGRATION EXECUTION ENVIRONMENT...................................88 26347 21012 21036 26700 EDISON EFTOS EIVIS ELTRAMOS 20170 977069 24404 22684 EMCP2 EMG-NET ENTICE EPOCH 23483 22503 EPSIS EROPPA 24758 ESCALATE 22509 20124 501853 ESD_NET ESOP EUDEM 21102 21040 26317 27741 EUROPORT-D EUROPRO EUROSTORE EUROTOOLS 22709 23770 23502 24903 FAST-SIM FINNOVA FITS FLASH 23480 FLEXSECURE 20082 20111 FRONTIER FSI-SD 26255 GABBY ANIMATOR GABBY)ANIMATOR : HOW TO RISE THE PRODUCTIVITY IN THE ANIMATION MARKET ....................................................................................................................... 132 GEODE GROUND EXPLOSIVE ORDNANCE DETECTION SYSTEM-MULTISENSOR FUSION SYSTEM FOR APL DETECTION LOCALIZATION AND CLASSIFICATION.............144 -EEUROPEAN ADVANCED DISTRIBUTION SIMULATION ENVIRONMENT.............146 EMBEDDED FAULT TOLERANT SUPERCOMPUTING................................................29 EMBEDDED INTERACTIVE VIDEO SERVER...............................................................33 ELECTRONIC MARKET AND TRADING OF MODELLING SERVICES AND ASSETS OF ENGINEERING SMES AND INSTITUTES.............................................................. 150 ELECTROMAGNETIC COMPATIBILITY USING PARALLEL PARAMETERISATION15 RESEARCH NETWORK FOR INTELLIGENT SUPPORT OF EMG STUDIES..............210 EUROPEAN NEW TECHNOLOGY FOR INDUSTRIAL & COMMERCIAL END-USERS95 ELECTROMAGNETIC PRODUCT OPTIMISATION USING COMPUTERS WITH HIGH PERFORMANCE..............................................................................................................63 REAL-TIME ANALYSIS OF VIDEO IMAGE STRUCTURE...........................................86 EXPERIMENTS ON REMOTE OPERATIONS FOR POST PRODUCTION APPLICATIONS...............................................................................................................52 EXPLOITATION OF SYNERGIES AND CO-OPERATIVE ACTIONS LEADING TO ACCELERATED TAKE-UP AND EFFECTIVENESS .................................................... 112 HIGH PERFOMANCE ELECTRONIC SYSTEMS DESIGN NETWORK.........................56 A PARALLEL OPTIMISATION SCHEME APPLIED TO EXTRUSION DIE DESIGN....12 STUDY ON THE STATE OF THE ART IN THE EU RELATED TO HUMANITARIAN DEMINING TECHNOLOGY, PRODUCTS AND PRACTICE........................................207 EUROPORT DISSEMINATION AND DEMONSTRATION.............................................50 EUROPEAN PROCESSING CORE TECHNOLOGY (EUROPRO)...................................45 A HIGH PERFORMANCE STORAGE PROJECT........................................................... 141 WORKING GROUP FOR EUROPEAN HPCN TOOLS PROMOTION........................... 154 -FMMP SYSTEM FOR SIMULATED NMR DATA ANALYSIS..........................................72 NOVEL APPROACHES TO FINNISH INDUSTRIAL COMPUTING...............................92 FORTRAN INTEGRATED TOOL SET.............................................................................89 HPCN TOOLS FOR ENHANCED HYDRODYNAMIC DESIGN OF FAST SHIPS ON PARALLEL COMPUTING PLATFORMS......................................................................115 MODULAR, SCALABLE DETECTION SYSTEM FOR EXPLOSIVES, DRUGS AND WEAPONS IN HAND BAGGAGE AND MEDIUM SIZED PARCELS.............................85 OPEN SYSTEM FOR COLLABORTIVE DESIGN.............................................................6 FLUID STRUCTURE INTERACTION- HIGH PERFORMANCE SIMULATION FOR STRUCTURAL DESIGN....................................................................................................8 -G26337 -H22696 24897 HAT HCHLOUSO 28283 HEATMASTER 21089 24003 25000 28249 22723 HIPERTEX HIPERTTN HIPSID HISTRIDE HITERM 20183 HOISE NM Octobre 1998 HIGH PERFORMANCE ARBITRAGE DETECTION&TRADING...................................69 HYDROCARBON AND CHEMICAL LOGISTICS OPTIMIZATION UNDER UNCERTAINTY VIA STOCHASTIC OPTIMIZATION.................................................113 INDUCTION HEATING MASTERING FOR METAL PROCESSING THROUGH INVERSE ANALYSIS....................................................................................................164 HIGH PERFORMANCE SYSTEM FOR TEXTILE EVALUATION..................................49 HIPERCOSME TTN.........................................................................................................93 HIGH PERFORMANCE SIMULATION FOR INTERACTIVE DESIGN........................ 125 A HIGH PERFORMANCE STRUCTURAL IDENTIFICATION ENVIRONEMENT......163 HIGH-PERFORMANCE COMPUTING AND NETWORKING FOR TECHNOLOGICAL AND ENVIRONMENTAL RISK MANAGEMENT ..........................................................75 HPCN EUROPE ON LINE NEWSMAGAZINE................................................................16 High Performance Computing and Networking / 218 HAND-HELD OPERATIONAL DE-MINING SYSTEM............... 183 29870 24900 21111 20074 HOPE HPCN-SCAT HPC-STANDARDS HPC-VAO 29737 28372 22707 20184 22693 HPGIN HPNURSA HP-RETAIL HPS-ICE HYPERBANK 24621 24949 22745 29944 ICETACT IMPACT INDEX INFIELD IRISH CENTRE FOR PROACTIVE TRANSFER OF ADVANCED COMPUTING.........105 INVERSE METHODS FOR WAVE PROPAGATION APPLICATIONS IN TIME DOMAIN118 INTELLIGENT DATA EXTRACTION.............................................................................78 24484 21017 INNO INNOVA 20216 INSIDE 21007 IOTA 20255 ISIS AN INSPECTION TECHNOLOGY TRANSFER NODE.................................................102 HIGH PERFORMANCE TELECONTROL STATION WITH IMAGE AND SOUND CAPABILITIES ................................................................................................................30 INTEGRATED SIMULATION AND DESIGN SYSTEM FOR CIVIL AND STRUCTURAL ENGINEERING................................................................................................................19 IOTA-INTELLIGENT OBJECT TRACKING AND ANALYSIS USING SMART SENSOR TECHNOLOGY................................................................................................................28 INTERACTIVE SATELLITE IMAGE SERVER ON GENERATION IMAGE CODING...22 20956 JAMES-RN 25050 JULIUS 28842 KLEE&CO 29255 977113 KLIF KNIXMAS 28928 KNOWNET 29812 LOTUS LIGHT ORDNANCE DETECTION BY TELEOPERATED UNMANNED SYSTEM................................................................. 182 26287 29479 M3D MACADAM 28363 22746 MADS MEDUSA 20118 MEMIPS 20089 MERCURY 22685 MESODYN 24410 24406 29909 20966 501852 26749 26293 METIER1 METIER2 METODIS MICA MIMEVA MIMIC MINEREC MULTI-SITE COOPERATIVE 3D DESIGN SYSTEM FOR ARCHITECTURE..............138 MULTISENSOR ACQUISITION CAMPAIGN FOR ANALYSIS AND DATA FUSION OF ANTIPERSONAL MINESS............................................................................................ 179 MULTI-PURPOSE ANTENNA DESIGN SIMULATOR.................................................167 MULTI-DISCIPLINARY ENGINEERING DESIGN VIA UNITARY SOFTWARE APPLICATIONS...............................................................................................................79 MODEL ENHANCED SOLUTION METHODS FOR INTEGER PROGRAMMING SOFTWARE .....................................................................................................................11 PERFORMANCE MANAGEMENT OF COMMERCIAL PARALLEL DATABASE SYSTEMS ..........................................................................................................................7 SOFTWARE FOR MESOSCALE DYNAMICS SIMULATIONS ON HPCN PLATFORMS IN CHEMICAL ENGINEERING.......................................................................................64 MECHANISM FOR ENABLING HPCN TECHNOLOGY TRANSFER IN EUROPE 1.....97 MECHANISM FOR ENABLING HPCN TECHNOLOGY TRANSFER IN EUROPE 2.....96 METACOMPUTING TOOLS FOR DISTRIBUTED SYSTEMS.... 189 MODEL FOR INDUSTRIAL CFD APPLICATIONS.........................................................27 MINE IMITATIONS FOR IN-FIELD EVALUATION..................................................... 206 MINING THE INTERNET FOR MARKETING INTELLIGENCE..................................151 DEMONSTRATION OF REAL TIME MINE RECOGNITION PROCEDURES ANALYSING GPR DATA ON EUROPRO.....................................................................139 29958 MINESEYE HPCN STOCHASTIC CORRELATION OF ANALYSIS AND TEST.............................. 114 HPC-STANDARDS ..........................................................................................................51 HIGH PERFORMANCE COMPUTATIONAL ENVIRONMENT FOR VIBRO-ACOUSTIC OPTIMIZATION ................................................................................................................5 HIGH PERFORMANCE GIGABIT I2O NETWORKING SOFTWARE........................... 181 HIGH PERFORMANCE NUMERICAL UNSTEADY ROTOR STATOR ANALYSIS....168 HIGH PERFORMANCE PARALLEL COMPUTING FOR MEDIUM SIZED RETAILERS71 HIGH PERFORMANCE SIMULATION OF INTERNAL COMBUSTION ENGINES.......17 HIGH PERFORMANCE BANKING.................................................................................67 -IINDUSTRY NGO FIELDTEST OF IMPROVED EFFICIENCY MULTISENSOR LANDMINE DETECTOR.................................. 190 -JINTERCONNECTION OF EUROPEAN RESEARCH AND UNIVERSITY NETWORKS AT 34-155 MBIT/S ...........................................................................................................24 JOINT INDUSTRIAL INTERFACE FOR END-USERS SIMULATIONS........................ 129 -KKNOWLEDGE AND LEARNINGS ENVIRONMENTS FOR EUROPEAN AND CREATIVE ORGANIZATIONS ..................................................................................... 172 A KNOWLEDGE AND LEARNING INFRASTRUCTURE.............................................177 KNOWLEDGE SHARED XPS-BASED RESEARCH NETWORK USING MULTI-AGENT SYSTEMS ...................................................................................................................... 211 KNOWLEDGE MANAGEMENT WITH INTRANET TECHNOLOGIES ....................... 173 -L-M- Octobre 1998 DEVELOPMENT AND OPTIMIZATION OF A DUAL SENSOR SYSTEM WITH REAL TIME DIGITAL SIGNAL PROCESSING FOR THE DETECTION AND IDENTIFICATION OF BURIED LANDMINES AND UNEXP.......................................................... 191 High Performance Computing and Networking / 219 501032 501031 MINESIGN MINETEST 26329 28942 22508 MISSIMU MOST MSC MEASUREMENT CAMPAIGN IN SUPPORT OF HUMANITARIAN DEMINING.......200 SUPPORT ACTIVITIES TO ESPRIT R&D PROJECTS RELATING TO HUMANITARIAN DEMINING .................................................................................................................... 199 MINIMALLY INVASIVE SURGICAL SIMULATOR.................................................... 142 MEDICAL REPRESENTATIVE ON-LINE SIMULATION TRAINING.......................... 175 MULTI-SITE COMPUTING.............................................................................................55 -N29907 NEPHEW NETWORK OF PC'S HETEROGENEOUS WINDOWS-NT ENGINEERING TOOLSET ........................................................... 187 22835 24414 NETAPDEX NOT SO MAD 26342 NOVICE NETWORK APPLICATION AND DISSEMINATION EXPERTISE.................................82 NODO DI TRANSFERIMIENTO TECNOLOGICO A SISTEMI DI OPERATORI DEL MANIFATTURIERO E DEI SERVIZI..............................................................................98 NETWORK ORIENTED VISUALIZATION IN A CLINICAL ENVIRONMENT............145 20116 24950 22717 ODESIM OPERA OPTIMUM OPTIMUM DESIGN OF MULTIBODY SYSTEMS..........................................................10 OPERATORS TRAINING DISTRIBUTED REAL-TIMESIMULATIONS ..................... 120 APPLYING HPC TECHNIQUES TO THE INDUSTRIAL DESIGN : AUTOMATIC OPTIMIZATION OF FORGING PROCESSES.................................................................74 21028 21037 22740 PACHA PACOS 2 PARMAT 20115 24960 24727 20162 23486 29895 20018 PAROS PARROT PDC TTN PHAROS PHASE PICE PINEAPL 26356 24724 20189 POEM PROHPC PROMENVIR 85386 PROMISE PARALLEL SCALABLE COMPUTERS FOR HIGH PERFORMANCE...........................32 PARALLEL COMPUTING FOR SPAIN 2........................................................................34 EFFICIENT HANDLING OF LARGE MATRICES ON HIGH PARALLEL COMPUTER SYSTEMS WITH THE PERMAS CODE..........................................................................77 PARALLEL LARGE SCALE AUTOMATIC SCHEDULING..............................................9 PARALLEL CREW ROSTERING ..................................................................................122 SWEDISH HPCN TECHNOLOGY TRANSFER NODE AT PDC...................................108 OPEN HPF PROGRAMMING ENVIRONMENT..............................................................14 A DISTRIBUTED PHARMACEUTICAL APPLICATIONS SERVER...............................87 PINPOINT, IDENTIFICATION, CLEARANCE AND ENSURANCE185 PARALLEL INDUSTRIAL NUMERICAL APPLICATIONS AND PORTABLE LIBRARIES........................................................................................................................4 PARALLEL PROCESSING OF VOLUMINOUS EDIFACT DOCUMENTS................... 147 PROMOTION OF HIGH PERFORMANCE COMPUTING.............................................107 HIGH PERFORMANCE COMPUTER BASED PROBALISTIC MECHANICAL DESIGN ENVIROMENT ................................................................................................................18 DEMONSTRATION AND ASSESSMENT OF HPCN IN MEDICAL IMAGING............192 25047 QUASI QUANTUM SIMULATION IN INDUSTRY...................................................................128 23475 RADAR 85387 RAIN 28245 RAMSES 22519 28410 RCNET ROXY HIGH PERFORMANCE RAINFALL RADAR IMAGE PROCESSING FOR SEWAGE SYSTEMS CONTROL......................................................................................................84 DEMONSTRATION AND ASSESSMENT OF HPCN IN NEURAL NETWORK APPLICATIONS FOR INDUSTRY AND MEDICINE.................................................... 193 REGIONAL EARTH OBSERVATION APPLICATION FOR MEDITERRANEAN SEA EMERGENCY SURVEILLANCE................................................................................... 161 EXPLOITING HPCN IN AN ENGINEERING CONSULTANCY ENVIRONMENT.........58 THE INTERNET LIVE INITIATIVE..............................................................................169 26267 501854 22691 SCHUMANN SEARCH² SEEDS 26276 21050 501851 22689 SEP-TOOLS SHAROC SIGEX SIMTRAP 23174 SISCI 22695 SLOEGAT 501033 SOFT-IT 24977 STRETCH -O- -P- -Q-R- -S- Octobre 1998 SUPPLY CHAIN UNCERTAINTY MANAGEMENT NETWORK OPTIMIZATION.....133 SEARCH FOR NEW TECHNOLOGIES FOR MINE SEARCH.......................................208 SIMULATION ENVIRONMENT FOR THE EVALUATION OF DISTRIBUTED TRAFFIC CONTROL SYSTEMS......................................................................................................66 AN INTER-OPERABLE SUITE OF EUROPEAN HPCN TOOLS...................................134 SYSTEM FOR HIGH-SPEED FOR AUTOMATIC READING OF CHEQUES..................47 SIGNATURE EXPLOITATION...................................................................................... 204 HPCN SIMULATION OF TRAFFIC INDUCED AIR POLLUTION USING PARALLEL COMPUTING IN A DISTRIBUTED NETWORK.............................................................65 STANDARD SOFTWARE INFRASTRUCTURE FOR SCI-BASED PARALLEL SYSTEMS ........................................................................................................................83 SHORT AND LONG TERM OPTIMIZATION OF ELECTRICITY GENERATION AND TRADING ........................................................................................................................68 SOFTWARE INTEROPERABILITY AND PLATFORM INDEPENDENCE: THE NEXT GENERATION OF SIMULATION ENVIRONMENTS................................................... 201 STORAGE AND RETRIEVAL BY CONTENT OF IMAGED DOCUMENTS.................123 High Performance Computing and Networking / 220 -T20957 24620 20231 24462 24428 23713 TEN-34 TETRAPC TOOLSHED TTNATGMD TTNMV-SF TTN-T TRANS-EUROPEAN NETWORK INTERCONNECT AT 34-155 MBPS..........................26 TECHNOLOGY TRANSFER IN PARALLEL COMPUTING.........................................104 TOOLS FOR HIGH PRODUCTIVITY ENGINEERING DESIGN.....................................20 TECHNOLOGY TRANSFER NODE AT GMD .............................................................. 101 TECHNOLOGY TRANSFER NODE OF MACHINE VISION IN FINLAND....................99 TECHNOLOGY TRANSFER NODE THURINGIA...........................................................91 24916 26286 28247 22517 VICAR VINE VISIT VISTA 85391 VR FOR EUROPE VIDEO INDEXING CLASSIFICATION ANNOTATION AND RETRIEVAL.................117 VIDEO ON DEMAND AND INTERACTIVE VIDEO EDITING SYSTEM..................... 137 VIRTUAL INTUITIVE SIMULATION TESTBED......................................................... 162 VIRTUAL INTERACTIVE STUDIO TELEVISION APPLICATION USING NETWORKED GRAPHICAL SUPERCOMPUTERS........................................................57 .......................................................................................................................................197 22505 WARPSPEID 22727 23516 85388 WEPTEL WINPAR -V- -W- 85389 85390 85392 501034 Octobre 1998 WIDE AREA PROJECT SUPPORTING PAN-EUROPEAN INFORMATION DISSEMINATION............................................................................................................53 INNOVATIVE WEATHER PRESENTATION ON TELEVISION.....................................76 WINDOWS BASED PARALLEL COMPUTING..............................................................90 DEMONSTRATION AND ASSESSMENT OF HPCN TOOLS FOR MESH GENERATION AND MESH PARTITIONING ........................................................................................ 194 DEMONSTRATION AND ASSESSMENT OF SCALABLE LOW-COST HPCN SYSTEMS FOR SIMULATION APPLICATIONS IN NON-TRADITIONAL INDUSTRIAL SECTORS195 DEMONSTRATION AND ASSESSMENT OF HPCN FOR INFORMATION MANAGEMENT AND DECISION SUPPORT APPLICATIONS....................................196 DEMONSTRATION AND ASSESSMENT OF HPCN IN MULTI-SITE APPLICATIONS FOR VIRTUAL REALITY.............................................................................................. 198 CLUSTER COMPUTING FOR DATA-INTENSIVE APPLICATIONS........................... 202 High Performance Computing and Networking / 221