Pervasive DataRush

This blog is syndicated from the Pervasive DataRush site.

August 2009 - Posts

  • How Accurate is Your Performance?

    Following up on our submission to the Netflix Challenge, the DataRush team is improving on our entry and tuning our Collaborative Filter into an application that you can use for your own "recommender" system.  Netflix offered a hefty million dollar prize to the team that could improve recommending appropriate movies to viewers by 10% or more.  Needless to say, we were motivated. 

    Using the Pervasive DataRush engine, our original final entry had an acceptable 6.5% improvement in accuracy over the course of the challenge with amazing performance.  We could crank through the full Netflix data in less than 17 minutes where other teams were taking a few hours.  Since the challenge, our DataRush team has been making some improvements.  Our performance is still orders of magnitude faster than the other submission and we are using a commodity 16-core system (too bad they weren't looking for performance speed alone!).   We have improved accuracy nearly 22% (8.3% vs. 6.5%) since the challenge with room for growth on accuracy and performance.  These accuracy results would have put us in 34th place out of thousands of submissions in the Challenge. 

    Our soon-to-be-released updated Collaborative Filter is based off DataRush 4.2 and brings a set of improvements, including accuracy and performance.  This is big news for E-tailers looking to recommend better products to their customers, like Amazon.  Or companion sites, like e-harmony, to shoot a straighter arrow.  The possibilities are endless... 

    Stay tuned because both DR4.2 and the Collaborative Filter should be released by the end of the year.

    www.pervasivedatarush.com

  • Ease of Use with Multicore Adoption

    Are you ready for the Multicore Revolution?  Many organizations are finding themselves weighed down and aren’t utilizing their commodity hardware to reach its full potential.  Nor is software scaling to exploit the full impact of the multicore processing power.  Programmers today need specialized knowledge in threading, concurrent memory access, deadlock detection, data workload partitioning and other complex aspects of parallel thread execution….Until now.    

     

    Jim Falgout, Chief Technologist for Pervasive DataRush, talks with an AMD spokesperson at Community One about the challenges facing companies considering adoption of multicore technology and the advantages of employing Java-based Pervasive DataRush engine, including ease of use and reduced design time. 

     

    Jim Falgout speaks with AMD at JavaOne about Multicore Adoption 

    View the complete You Tube video:  http://www.youtube.com/watch?v=US-AL1ugaIc

     

     

     

     

     

More Posts