Date of Award

5-20-2011

Degree Type

Dissertation

Degree Name

Ph.D.

Degree Program

Engineering and Applied Science

Department

Computer Science

Major Professor

Abdelguerfi, Mahdi

Second Advisor

Tu, Shengru

Third Advisor

Richard, Golden

Fourth Advisor

Chen, Huimin

Fifth Advisor

Sample, John

Abstract

The goal of this research is to provide a supporting Web services architecture, consisting of a service model and catalog, to allow discovery and automatic orchestration of geospatial Web services. First, a methodology for supporting geospatial Web services with existing orchestration tools is presented. Geospatial services are automatically translated into SOAP/WSDL services by a portable service wrapper. Their data layers are exposed as atomic functions while WSDL extensions provide syntactic metadata. Compliant services are modeled using the descriptive logic capabilities of the Ontology Language for the Web (OWL). The resulting geospatial service model has a number of functions. It provides a basic taxonomy of geospatial Web services that is useful for templating service compositions. It also contains the necessary annotations to allow discovery of services. Importantly, the model defines a number of logical relationships between its internal concepts which allow inconsistency detection for the model as a whole and for individual service instances as they are added to the catalog. These logical relationships have the additional benefit of supporting automatic classification of geospatial services individuals when they are added to the service catalog. The geospatial service catalog is backed by the descriptive logic model. It supports queries which are more complex that those available using standard relational data models, such as the capability to query using concept hierarchies. An example orchestration system demonstrates the use of the geospatial service catalog for query evaluation in an automatic orchestration system (both fully and semi-automatic orchestration). Computational complexity analysis and experimental performance analysis identify potential performance problems in the geospatial service catalog. Solutions to these performance issues are presented in the form of partitioning service instance realization, low cost pre-filtering of service instances, and pre-processing realization. The resulting model and catalog provide an architecture to support automatic orchestration capable of complementing the multiple service composition algorithms that currently exist. Importantly, the geospatial service model and catalog go beyond simply supporting orchestration systems. By providing a general solution to the modeling and discovery of geospatial Web services they are useful in any geospastial Web service enterprise.

Rights

The University of New Orleans and its agents retain the non-exclusive license to archive and make accessible this dissertation or thesis in whole or in part in all forms of media, now or hereafter known. The author retains all other ownership rights to the copyright of the thesis or dissertation.

Share

COinS