Shortcut selection in RDF databases

Vicky Dritsou, Panos Constantopoulos, Antonios Deligiannakis, Yannis Kotidis

Abstract: An increasing amount of data produced nowadays is in RDF format. While significant work has been performed on view selection and query optimization algorithms in relational database systems, little attention has been paid to the problem of optimizing the performance of query workloads in RDF databases. In this paper we propose the notion of shortcuts, as a method for reducing the query processing cost. We then devise a greedy algorithm that, given a space constraint and a query workload in an RDF database, seeks to select the optimal shortcuts to materialize. Our experiments validate our approach and demonstrate that our algorithm manages to significantly reduce the query processing cost while keeping space requirements low.