SpiderStore: Exploiting Main Memory for Efficient RDF Graph Representation and Fast Querying
Robert Binna, Eva Zangerle, Dominic Pacher · 2010
The constant growth of available RDF data requires fast and efficient querying facilities of graph data. So far, such data sets have been stored by using mapping techniques from graph structures to relational models, secondary memory structures or even complex main memory based models. We present the main memory database SpiderStore which is capable of efficiently managing large RDF data sets and providing powerful and fast SPARQL processing facilities. The SpiderStore storage concept aims at storing the graph structure in main memory without performing any complex mappings. Therefore it exploits the natural web-structure of RDF by using fast and random access to main memory. The abandonment of additional mappings or meta-information therefore leads to a significant performance gain compared to other common RDF stores.