Evaluating graph traversal algorithms for distributed SPARQL query optimization
JIST'11 Proceedings of the 2011 joint international conference on The Semantic Web
| Hi-index | 0.00 |
Here we describe the application of two well knowngraph algorithms, Edmonds' algorithm and Prim'salgorithm, to the problem of optimizing distributedSPARQL queries. In the context of this paper, a“distributed SPARQL query” is a SPARQL querywhich is resolved by contacting any number of remoteSPARQL endpoints. Two optimization approaches aredescribed. In the first approach, a static query plan iscomputed in advance of query execution, using one oftwo standard graph algorithms for finding minimumspanning trees (Edmonds' algorithm and Prim'salgorithm). In the second approach, the planning andexecution of the query are interleaved, so that as eachpotential solution is expanded it is permitted to followan independent query plan. Our optimization approachrequires basic statistics regarding RDF predicateswhich must be obtained prior to the user's query,through automated querying of the remote SPARQLendpoints.