The work package of the Austrian Grid project focuses on retrieval, processing and managing of heterogeneous data stored in distributed databases. These distributed databases are organized by using GRID technology and methods. In a GRID environment there is no need to know where particular information is stored. Therefore data retrieval, processing and management have its own demands.
To facilitate data retrieval of heterogeneous data from the GRID it is necessary to use a mechanism that is aware of the diverse data structures. Each part of these data structures is defined by a unique schema. It is necessary to merge the local data schemas into a more global data schema, in order to provide a general ubiquitous view of the relevant information.
In general such a (global) data schema has to cope with semantic conflicts between the referenced local data schemas. Ontologies are the state-of-the-art technology to dissolve these semantic conflicts. Ontologies allow the representation of semantically enriched data (e.g. in a local data base). Ontology mapping enables to bridge the semantic gap between heterogeneous and distributed data sources. These semantic bridges are a prerequisite for data combination and consolidation between distributed data bases independent from using them between a local and a global data schema or between local data schemas directly.
The ontology approach applied to environments associated with GRID computing solves the main problem for data retrieval and processing on an application layer. Therefore the WP provides the necessary data retrieval mechanisms for the Environmental GRID application (in cooperation with other WPs).
The described work package is part of the Austrian Grid Project, funded by the Austrian BMBWK (Federal Ministry for Education, Science and Culture) under contract GZ 4003/2-VI/4c/2004.