Towards content-aware SPARQL query caching for semantic web applications

Abstract

Applications are increasingly using triple stores as persistence backends, and accessing large amounts of data through SPARQL endpoints. To improve query performance, this paper presents an approach that reuses results of cached queries in a content-aware way for answering subsequent queries. With a focus on a common class of conjunctive SPARQL queries with filter conditions, not only does the paper provide an efficient method for testing whether a query can be evaluated on the result of a cached query, but it also shows how to evaluate the query. Experimental results show the effectiveness of the approach. © 2013 Springer-Verlag.