|
Abstract : |
Most applications of in database mining at present seem to be concentrated around the discovery of knowledge for application-oriented decision support. In this paper we describe a tool for enhancing the performance of the database management system itself using semantics captured by database mining techniques. We investigate the use of database mining in supporting state-aware query optimization where improved performance is sought by dynamically increasing the utilization of available resources. We consider, in particular, the use of semantic knowledge in enhancing the system performance and the utilization of specialized hardware such as ICL's SCAFS by semantic query reformulation. The idea is to discover semantic information about the data stored in the database using database mining techniques and use it to reformulate queries to make better use of available resources at the time of execution of the query. Thus the execution strategy of the query is based on the current state of the system (and so 'state-aware'). To achieve this we create a rewrite system, consisting of rules and equations mined from the database, that rewrites queries made to the database for state-aware execution. We provide an architecture for this query pre-processor and discuss performance considerations involved in its implementation., |
