4-0-0-4 MBICBI802: HIGH PERFORMANCE COMPUTING Course objective: The course will equip the students with essential concepts of High Performance Computing and the importance of Cloud Computing architectures in today’s world. UNIT - I 15 periods PARALLEL PROCESSING AND PIPELINING PROCESSING Overview of parallel processing and pipelining processing, study and comparison of uniprocessors and parallel processors, conventional and EPIC architecture; Evolution of parallel processors, future trends and there architecture. Necessity of high performance, Constraints of conventional architecture, Parallelism in uniprocessor system, Evolution of parallel processors, future trends, Architectural Classification, Applications of parallel processing, Instruction level Parallelism and Thread Level Parallelism, Explicitly Parallel Instruction Computing (EPIC) Architecture, Case study of Intel Itanium Processor. Principles of scalable performance: Performance Metrics and Measures, Speedup Performance Laws. Programming aspects for Intel Itanium Processor UNIT - II 15 periods SIMD and ARRAY PROCESSORS Parallel computers- Taxonomy of parallel computing paradigms SIMD Computer Organisation Masking and Data network mechanism, Inter PE Communication, Interconnection networks of SIMD, Static Vs Dynamic network, cube hyper cube and Mesh Interconnection network. Parallel Algorithms for Array Processors: Matrix Multiplication. Sorting, SIMD computer organisation; Implementation issues of Matrix multiplication and sorting on array processor and their analysis, Distributed memory programming with MPI, Shared memory programming with openMP UNIT - III 15 periods CLOUD COMPUTING AND VIRTUALIZATION TECHNOLOGY Introduction to cloud computing, definition, architecture-characteristics, components, Cloud computing platforms, Saas, Paas, Iaas and others, Google App Engine- Amazon Ec2- Microsoft Azure, Cloud Deployment models are cloud types: community, hybrid, private, and public clouds. Virtualization technology- virtual machine technology- virtual machines and elastic computing- desktop virtualization and application streaming-automating infrastructure management UNIT - IV 15 periods DATE AND FILE SYSTEMS IN CLOUD Data in the cloud: Relational databases, Cloud file systems: GFS and Hadoop File System, BigTable, HBase and Dynamo. Map-Reduce and extensions: Parallel computing, The mapReduce model, Parallel efficiency of Map-Reduce, Relational operations using Map-Reduce, Features and comparisons among GFS, HDFS REFERENCES 1. “Cloud Computing : A Practical Approach”, Anthony T. Velte Toby J. Velte, Robert Elsenpeter, The McGraw-Hill Osborne, 2009, 1e. 2. “Cloud Computing Bible”, Barrie Sosinsky, John Wiley & Sons, 2011, 1e. LEARNING OUTCOMES UNIT – I UNIT – II UNIT – III UNIT –IV The learner will be able to understand the essentials of parallel processing in computing architectures The learner will be able to explore the parallel computing and its types and various algorithms associated with it. The learner will understand the importance of Cloud computing concepts and the various deployment models and service oriented architectures The learner will be able to gain the knowledge on Cloud computing file systems and Map reduce algorithms for parallel computing