Super-Fast RDF Search Engine Developed

The Register is reporting that Irish researchers have developed a new high-speed RDF search engine capable of answering search queries with more than seven billion RDF statements in mere fractions of a second. "'The importance of this breakthrough cannot be overestimated,' said Professor Stefan Decker, director of DERI. 'These results enable us to create web search engines that really deliver answers instead of links. The technology also allows us to combine information from the web, for example the engine can list all partnerships of a company even if there is no single web page that lists all of them.'"

Comments

Next up: Ontology spam

[G4from128k]

Yes, creating a consistent ontology is challenge. But the bigger challenge is the lack of incentive for ontology truthfulness. If this type of search becomes popular, ontology spam and OSEO (Ontology Search Engine Optimization) will become a booming industry.

Could be interesting, but missing details

[Anonymous Coward]

What kind of data set did they use? The structure and contents of the graph that is the data in an RDF database has a huge impact on the performance of query execution, and different applications have different structures.

What kind of queries are they running? There are several different RDF query languages (think of SeRQL, RDQL, N3, SPARQL, etcetera) and some of them support quite complex queries. Quickly finding the answers to a simple query like

SELECT ?name WHERE ?name <http://xmlns.com/foaf/0.1/name> "John Smith"

is just a matter of an indexed lookup and not very special. But, like in SQL, much more complex expressions can be generated that require complex index operations on the query execution level. Having implemented an RDF database that supports SPARQL queries an order of magnitude faster than the software the W3C uses for their experiments (which, admitedly, doesn't have performance as a prime requirement), I know that it's possible to do simple things fast, but the interesting part is handling RDF queries that don't easily map to efficient database operations.

Which brings me to the most important point: where is their detailed report? Can I get the software somewhere and perform my own tests? The article is too vague to draw any conclusions about what their RDF database does, and how good it is. I'd love to read up on it, but I can't seem to find the information.

Here's the Tech Report

[aharth]

Hello, I am one of the main developers of SWSE. True, the press release is vague, but there is only so much you can say in a press release aimed for the general public.

We have a Technical Report available at http://www.deri.ie/fileadmin/documents/DERI-TR-200 7-04-20.pdf that should answer most of the technical questions.

From the abstract:

"We present the architecture of an end-to-end search engine that uses a graph data model to enable interactive query answering over structured and interlinked data collected from many disparate sources on the Web.

In particular, we study distributed indexing methods for graph-structured data and parallel query evaluation methods on a cluster of computers.

We evaluate the system on a dataset with 430 million statements collected from the Web, and provide scale-up experiments on 7 billion synthetically generated statements."