| BitMat: A Main-memory Bit Matrix of RDF Triples for Conjunctive Triple Pattern Queries (2009) | |||||||||||||
Abstract | |||||||||||||
| This poster proposes BitMat, a bit matrix structure for representing a large number of RDF triples in memory and processing conjunctive triple pattern (multi-join) queries using it. The compact in-memory storage and use of bitwise operations, can lead to a faster processing of join queries when compared to the conventional RDF triple stores. Unlike conventional RDF triple stores, where the size of the intermediate join results can grow very large, our BitMat based multijoin algorithm ensures that the intermediate result set remains small across any number of join operations (provided there are no Cartesian joins). We present the key concepts of BitMat structure, its use in processing join queries, describe the preliminary experimental results with UniProt and LUBM datasets, and discuss the possible use case scenarios. | |||||||||||||
Publication details | |||||||||||||
| |||||||||||||