SPIDER: a system for scalable, parallel / distributed evaluation of large-scale RDF data