FTHPIS

Fault Tolerant High Performance Information System (FTHPIS) is a system designed to provide dynamic publishing and discovering of web service information for large scale grid projects. The system suggests Gaggle of Information Services where each Information Service is responsible of providing both publishing and discovering of services and their contextual information (WS-Context).

Services, because they are generic, must provide additional, descriptive metadata in order to be useful. The problem is simple: a client may interact with two different Services in exactly the same way (the WSDL is the same), but these Services may hold different data. Open GIS Consortium (OGC) compatible data services (Web Feature Services) are good examples for this. One Web Feature Service, for example, may contain GPS data for the Western United States while the other has GPS data for Northern Japan.  Clients must be able to query information services that encode (in standard formats) all the necessary information, or metadata, that enables the client to connect to the desired service.  This is an example of the very general problem of managing information about Web Services. To address these problems, we are designing a general purpose information system, the Fault Tolerant High Performance Information System (FTHPIS), that we are applying initially to problems in GIS information management .

We use the Universal Description, Discovery, and Integration (UDDI) (Bellwood, Clement, and von Riegen, 2003) specifications in our design as centralized registry. UDDI offers users a unified and systematic way to find service providers through a centralized registry of services. We design an extension to existing UDDI Specifications in order to provide dynamically updated service registry data. Existing UDDI Information Model [.ppt]. Proposed UDDI Information Model [.ppt]

Here is the schema for proposed extensions to UDDI v3. in order to annotate service descriptions with more information. We introduce an extensions for both metadata about a service and life time of a service. You may find the image for proposed metadata schema at [.jpg]. You may also find proposed schema for lifetime information about a service at [.jpg].

[1] - iSERVO: Implementing the International Solid Earth Research Virtual Observatory by Integrating Computational Grid and Geographical Information Web Services, To be published in Special Issue of Pure and Applied Geophysics (PAGEOPH) [.pdf]

• OGC Service Taxonomy [.xml]
• Service Taxonomy based on Service Management [.xml]
• A preliminary Taxonomy to categorize WS metadata [.xml]
• OGC Quad Taxonomy [.xml]

An information mall for Community Grids Lab Activities is located at [html]. You may access Community Grids Lab publications at following link [html]. Also, Community Grids Lab presentations are available at [html]. Concurrency and Computation: Practice and Experience publications are available at [html].

This work is supported by the NASA Advanced Information Systems Technology (AIST) program.

Mehmet S. Aktas
CGL Showers Building Suite 224,
Bloomington, Indiana 47408
Email:maktas@cs.indiana.edu

Dr.Marlon Pierce
CGL Showers Building Suite 222,
Bloomington, Indiana 47408
Email:marpierc@indiana.edu

Dr.Geoffrey Fox
CGL Showes Building Suite 222,
Bloomington, Indiana 47408
Email:gcf@indiana.edu