Seismic Data Portal
Portal Staff:
EMSC
Remy Bossu
Sergio Rives
Phetaphone Thomy
ORFEUS - KNMI
Torild Van Eck
Luca Trani
Alessandro Spinuso (myself)
In collaboration with:
ArcLink middleware
QuakeML specs
Edinburgh 06/11/2008
Seismic Data Portal
NERIES objective..
Networking together seismological observatories and research institutes into one
integrated European infrastructure that provides access to data and data products for
research...
..and approach
Design and develop a
cyberinfrastructure for distributed
and heterogeneous data streams
and services.
Design and develop a Web portal
that acts as the uppermost layer of
the infrastructure and provides
rendering capabilities for the
underlying sets of data.
Edinburgh 06/11/2008
Seismic Data Portal
Conceptual Layers
Data Levels
Raw Data (Level I)
Waveform data, accelerometric data,
Station location charateristics,
Theoretical data
Derived Data (Level II)
Seismic picks, amplitudes,
automatic Magnitudes
Moment Tensors
Interpreted Data (Level III)
Events (location, magnitude, etc)
Event maps, Shakemaps,
Loss estimation maps,
Newsletters, etc.
Edinburgh 06/11/2008
Seismic Data Portal
The idea of “Distributed data” includes the concept of
Distributed maintenance and responsibility which persist in time
Sounds quite obvious for the “External Software Agents” Layer (machine to machine
communication).
Can we apply this idea to
the presentation logic
layer? ( human machine communication)
Advantages
Rapid re-deployment of improved tools, statistics on the usage of the applications can be made at
the data provider place. Helps the organization of a distributed development team with different
responsibilities either on the data or on the tools. (NERIES DT structure)
Our Strategy
Data access through Webservices, Portal Frameworks, and WSRP for the presentation
logic layer of distributed webapplications
Edinburgh 06/11/2008
Seismic Data Portal
Software Architecture
Each single user application of the portal consists of a Java-based JSR-168 standard portlet (often
provided with interactive maps for data discovery)
All the portlets are
connected to the
underlying
webservices
infrastructure which
adopts appropiate
interfaces and
formats, wrapping the
middleware used by
the seismological
observatories or
institutes providing the
data
Edinburgh 06/11/2008
Seismic Data Portal
SeismoLink webservice: Stations inventory and waveform data retrieval.
Neries Portal
Wraps the whole ArcLink
network.
Single access to the
whole infrastructure
Seismolink WebService
getInventory
dataRequest
dataRetrieve
purgeData
checkStatus
arclink TCP client
ODC
arclink node
generic
arclink node
GFZ
arclink node
Edinburgh 06/11/2008
Seismic Data Portal
Local Portlets presentation logic
QuakeML Ws
SMI-RDF Ws
SOAP Calls
Local Portlet
- Spring Portlet MVC
DWR Calls
- Apache Jetspeed2 portal
server
- Google Maps API
- Ajax, DWR
Edinburgh 06/11/2008
Seismic Data Portal
Remote Portlets presentation logic
QuakeML Ws
SeismoLink Ws
- Spring Portlet MVC
SMI-RDF Ws
TauP arrival times Ws
Proxyed Calls
ResourceProxy Servlet
P
R
O
D
U
C
E
R
- Apache Pluto 1.0.2 container
with in house addons for
displaytag compatibility
C
O
N
S
U
M
E
R
- Apache Jetspeed2 portal
server
- Apache WSRP4J Producer
with in house addons for remote
username exchange
Ajax Rest/SOAP Calls
- WSRP Proxy Portlet to register
and access remote portlets
- ResourceProxy Servlet for
GET and POST http Ajax calls
- Google Map API
- Javascript Parasitic
Inheritance dev approach
Edinburgh 06/11/2008
Seismic Data Portal
User datasets selection exchange
User`s Events Selection
“Event Explorer”
Jetspeed Portlet
StoredUser Events Cart Ws
locally
Multi Modal
waveform requests,
Event or Time Modes
“Waveform Explorer”
Remote Portlet
Edinburgh 06/11/2008
Seismic Data Portal
QuakeML and RDF Repositories
The parametric information about
a seismic event is delivered
using a seismology-specific
Extensible mark-up Language
(XML) dialect called QuakeML
URIs are used to assign
identifiers either to seismic-event
parameters described by
QuakeML, or to generic
resources, for example,
authorities, locations providers,
location methods, software
adopted, and so on, described by
a data model constructed with
the resource description
framework (RDF).
smi URI schema
Both repositories are
accessible as webservices
either in SOAP or in RESTful
way
Edinburgh 06/11/2008
Seismic Data Portal
SMI (Seismological Meta Information)
Data Model
Resource
Class Type
the RDF model can be extended, within certain
constraints, with new classes and properties.
Edinburgh 06/11/2008
Seismic Data Portal
SMI Administrative tool (adapts to the RDF Model)
Resource
Class Type
Registered Webservice
Resources
Edinburgh 06/11/2008
Seismic Data Portal
Waveform Requests Cart
Dataset user
description
Request ID as URI.
The request`s info could be stored into the SMI Model
Link between Seismic
Event and waveforms
Edinburgh 06/11/2008
Seismic Data Portal
References
- The Portal technologies adopted are all part of the
project, that supports
standards such as JSR-168 for portlets develoment and WSRP 1.0 (WSRP4J)
http://portals.apache.org
- The QuakeML and SMI-RDF Webservices are designed to be consumed through a SOAP or a RESTful
access type, and together with the SeismoLink and User Data webservices, have all been implemented
using the Axis 2 framework
http://ws.apache.org/axis2/
- is a layered Java/J2EE application framework, based on code published in Expert One-on-One
J2EE Design and Development by Rod Johnson (Wrox, 2002).
http://www.springframework.org
- QuakeML is a flexible, extensible and modular XML representation of seismological data which is
intended to cover a broad range of fields of application in modern seismology.
https://quake.ethz.ch/quakeml
- The RDF specifications provide a lightweight ontology system to support the exchange of knowledge on
the Web. The SMI-RDF API have been developed thanks to the
http://jena.sourceforge.net/
http://www.w3.org/RDF/
-ArcLink is the distributed data archive access protocol. Following the simple but very efficient appoach of
SeedLink real-time communication protocol, [..] It supports access both to meta databases (stations
inventory) and waveform archives. Waveform utput is in SEED or MiniSEED format, meta data output in
XML or dataless SEED (response information only).
http://www.webdc.eu/webdc_sum.html
Edinburgh 06/11/2008