W3C home > Mailing lists > Public > www-rdf-interest@w3.org > October 2007

[announcement] BigOWLIM 0.9.6 loads UNIPROT and delivers superior query performance

From: Damyan Ognyanoff <damyan@sirma.bg>
Date: Thu, 25 Oct 2007 14:17:09 +0300
Message-ID: <B4A24F6E6EC24B91ACEEEA9D98CAE2DA@sirma.int>
To: "interested-in-OWLIM" <interested-in-OWLIM@ontotext.com>
Cc: <sesame-interest@lists.sourceforge.net>, <seweb-list@lists.deri.at>, <www-rdf-interest@w3.org>, <semantic-web@w3.org>, <tripcom@lists.deri.org>, <tao@dcs.shef.ac.uk>, <mediacampaign@iis-list.joanneum.ac.at>, <all@semantic-gov.org>


(*** We apologize if you receive multiple copies of this announcement ***)



BigOWLIM 0.9.6 loads UNIPROT and delivers superior query performance

 Ontotext released version 0.9.6 of BigOWLIM - its enterprise-class OWL engine. 


BigOWLIM remains the only engine which can perform non-trivial OWL reasoning with 1 billion statements. While in 0.9.6 the data loading performance was further improved, the emphasize of this release was on boosting the query performance to levels which allow multiple simultaneous users to perform real-time queries against huge datasets. 

The major changes in the new release can be summarized as follows:

.         Dramatically improved query evaluation speed through better optimizations:

-        Now BigOWLIM is faster than SwiftOWLIM, even on top of few million statements

-        About 100 times faster than BigOWLIM ver. 0.9.5 on LUBM(8000)

.         Loading became twice faster, compared ver. 0.9.4 and earlier

.         Updated distribution - setting up an application is much easier and faster (as in SwiftOWLIM 2.9)

.         Numerous fixes and corrections in response to customer feedback. 

Version 0.9.6 of BigOWLIM was the first engine which managed to load and perform inference (total materialization) against the full OWL representation of UNIPROT - the world's most comprehensive catalog of information about proteins and one of the largest and most popular life science databases. 

More details about the UNIPROT runs and query evaluation performance can be found at the end of the standard OWLIM presentation. Slides 9 and 10 there present a map of the most interesting published results on scalable reasoning, including data from AllegroGraph, Openlink Virtouso, and ORACLE 11g.



Damyan Ognyanoff,
Ontotext Lab, Sirma Group Corp.
http://www.ontotext.com
Received on Thursday, 25 October 2007 11:17:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:20 GMT