Automatic Generation of Warehouse Mediators Using an Ontology Engine

Terence Critchlow, Madhavan Ganesh, Ron Musick
Center for Applied Scientific Computing
Lawrence Livermore National Laboratory
P.O. Box 808, L-561, Livermore, CA 94551
{critchlow | ganesh | rmusick}@llnl.gov

Abstract

Data warehouses created for dynamic scientific environments, such as genetics, face significant challenges to their long-term feasibility. One of the most significant of these is the high frequency of schema evolution resulting from both technological advances and scientific insight. Failure to quickly incorporate these modifications will render the warehouse obsolete, yet each evolution requires significant effort to ensure the changes are correctly propagated. DataFoundry utilizes a mediated warehouse architecture with an ontology infrastructure to reduce the maintenance requirements of a warehouse. Among other things, the ontology is used as an information source for automatically generating mediators, the methods that transfer data between the data sources and the warehouse. The identification, definition and representation of the metadata required to perform this task is a primary contribution of this work.

Appeared in

Proceedings of the 5th KRDB Workshop, Seattle. WA. May 1998.

Look at the Paper (ps.gz, pdf.gz)