We describe an enterprise middleware system that integrates, from a user’s perspective, data located on disparate data storage devices without imposing additional requirements upon those storage mechanisms. The system provides advanced search capabilities by exploiting a repository of metadata that describes the integrated data. This search mechanism integrates information from a collection of XML metadata documents with diverse schema. Users construct queries using familiar search terms, and the enterprise system uses domain representations and vocabulary mappings to translate the user’s query, expanding the search to include other potentially relevant data. The enterprise architecture allows flexibility with respect to domain dependent processing of user data and metadata
Revised: August 17, 2010 |
Published: February 27, 2003
Citation
Thomson J., A.R. Chappell, and J.P. Almquist. 2003.Enterprise Middleware for Scientific Data. In Proceedings of the 5th International Conference on Enterprise Information Systems (ICEIS - 2003), edited by O Camp and JBL Fillipe, 112-119. Dordrecht:Kluwer.PNNL-SA-37901.