Grids of Grids and Service Oriented Architectures
When building a Grid of Grids [GOG] using Service Oriented Architecture [SOA] principles, we build families of services and integrate them with common messaging substrates and information/discovery services. Realistic simulations require that the computational services (“Execution Grid” services) be coupled to data sources (“Data Grid” services). Examples of simulation and modeling code geospatial data requirements include: locations and descriptions of input sensors; local, state, and national political boundaries; earth surface terrain features; transportation infrastructure including highways, railroads; demographics and census data; and infrastructure such as hospitals, police and fire stations, etc. All of these data types may be managed by Geographical Information System (GIS) services.
GIS Data Grids
When building the GIS Data Grid services, we must acknowledge and support the extensive previous work that has been done. An overview is available in [Sayar2004]. The Open Geospatial Consortium (OGC) [OGC] provides several important, open specifications for building GIS services that will allow us to leverage other groups’ efforts. Relevant OGC data model specifications include the Geographic Markup Language (GML) and the SensorML family [SensorML] of data and metadata information models. OGC service specifications include the Web Feature Service (WFS) [Vretanos2002], which provides access to archival data and abstract map features, and the Web Map Service (WMS) [Beaujardierre2004], which generates human-comprehensible overlay maps. A WMS may generate all images locally from abstract information stored in remote WFS instances, or it may combine its own image capabilities with maps from other WMS instances.
GIS Work at CGL
Over the last year, the Community Grids Lab has undertaken a comprehensive project to build GIS services using current Grid and Web Service standards. By following Web Service standards, we may incorporate general Web Service functionality, including security, reliability, and information management. In summary, we are implementing
* SensorML-based Sensor Grid services * Web Feature Service * Web Map Service
Research Challenges for GIS Data Grids
* SensorML specification family focuses primarily on sensor metadata (location, type of data, etc) and data formats. We still must supply an interesting implementation of Sensor Grid services that make use of SensorML information. These include information services for discovering and querying available sensors and the software infrastructure for getting real-time, streaming data to the simulation codes. As described elsewhere, the NaradaBrokering system provides an excellent substrate for managing both small, event-based messages as well as continuous data streams. * WFS: WFS provides a repository for GIS archival data. The specification provides the common framework, but we must still provide interesting implementations. High performance data transfer is one prominent challenge in scientific Grid applications in general and specifically for this project. We will investigate using the XML Infoset to provide highly efficient XML representations that should provide significant performance enhancements in current services. * WMS: Mapping tools are obviously useful for presenting information and results. Much of the data that we will work with includes both time-dependent and time-independent features, so we must design and build a streaming WMS that is capable of delivering these evolving features. Streaming imagery also will allow us to integrate GIS map services into our more general collaboration infrastructure (GlobalMMCS). This allows multiple participants in specific sessions to share maps, video, audio, white boards, and other collaboration tools.
[Sayar2004] Ahmet Sayar, Marlon Pierce, Geoffrey Fox OGC Compatible Geographical Information Services Technical Report (Mar 2005), Indiana Computer Science Report TR610.
[OGC] The Open Geospatial Consortium (OGC): http://www.opengeospatial.org
[SensorML] Sensor Model Language (SensorML) project and specification page: http://vast.nsstc.uah.edu/SensorML
[Beaujardierre2004] de La Beaujardiere, Jeff, Web Map Service, OGC project document reference number OGC 04-024.
[Vretanos2002] Vretanos, P (ed.) (2002), Web Feature Service Implementation Specification, OpenGIS project document: OGC 02-058, version 1.0.0.
[FTHPIS] Fault Tolerant High Performance Information System (FTHPIS) Project Web Site: http://grids.ucs.indiana.edu/~maktas/fthpis
[Cox2003] Cox, S., Daisey, P., Lake, R., Portele, C., and Whiteside, A. (eds) (2003), OpenGIS Geography Markup Language (GML) Implementation Specification. OpenGIS project
[Aktas, 2004] Aktas, M., Aydin, G., Donnellan, A., Fox, G., Granat, R., Grant, L., Lyzenga, G., McLeod, D., Pallickara, S., Parker, J., Pierce, M., Rundle, J., Sayar, A., and Tullis, T. iSERVO: Implementing the International Solid Earth Research Virtual Observatory by Integrating Computational Grid and Geographical Information Web Services Technical Report December 2004. Submitted to Special Issue of Pure and Applied Geophysics (PAGEOPH) for Beijing ACES Meeting July.
[Aydin2005] Galip Aydin, Progress Report on SCIGN GPS Sensor Grid Implementation. http://complexity.ucs.indiana.edu/~gaydin/sopac/summary.doc
[SOA] D. Booth, H. Haas, F. McCabe, E. Newcomer, M. Champion, C. Ferris, and D. Orchard, “Web Services Architecture.” W3C Working Group Note 11 February 2004. Available from http://www.w3.org/TR/2004/NOTE-ws-arch-20040211/
[GOG] Geoffrey Fox, Shrideep Pallickara, and Marlon Pierce Building a Grid of Grids: Messaging Substrates and Information Management to appear as chapter in book "Grid Computational Methods" Edited by M.P. Bekakos, G.A. Gravvanis and H.R. Arabnia. http://grids.ucs.indiana.edu/ptliupages/publications/GridOfGrids.pdf