ResourceDescriptionFramework

Triple Store Performance

Pasted here in haste...-- RogerHyam - 03 May 2006

I believe scalability and performance in RDF stores IS a major problem and I think we should try to gather as much information about it as possible. Roger, could we start to collect resources in the TAG wiki about it? So far I am aware of the following ± recent articles:

* http://crschmidt.net/blog/archives/90/rdf-as-backing-storage/ * http://simile.mit.edu/reports/stores/ * http://km.aifb.uni-karlsruhe.de/ws/psss03/proceedings/harris-et-al.pdf * other interesting papers from the same conference "Practical and Scalable Semantic Systems Service": http://km.aifb.uni-karlsruhe.de/ws/psss03/proceedings * http://eprints.ecs.soton.ac.uk/11985/01/owens-semanticStorageOverview-ecs.pdf * http://www.w3.org/2001/sw/Europe/reports/rdf_scalable_storage_report/

Most of them are sceptical about having large scale rdf stores in the intermediate future and usually suggest to go with a specialised stores for specific appllication purposes. Especially the combination of storing large amounts of data based on complex ontologies with good inference/reasoning/querying support seems unreasonable in a generic solution.

But maybe Oracle can do much better. Has anyone heard claims about their performance yet?

Markus 2006-04-19

• TechnologyRoundup