IRTG: A grid middleware for bioinformatics

Frequently used queries in bioinformatics applications should be given special treatment. This is because: (1) there are certain queries submitted by scientists and researchers over and over again, and (2) biological databases are usually too big and complicated to be rapidly queried every time a user submits a query. We propose in this study a query processor called IRTG that expedites the processing of frequently used queries. IRTG can be considered as a type of middleware running at a central site. It classifies queries per their types and ranks the query types based on their frequencies. The results of queries whose types are classified as frequently used types will be cached at the middleware. Thereafter, each frequently used query of these types will be answered from the middleware rather than from remote sites, which improves the query's response time. We evaluate IRTG experimentally.