benchmarkingblog

Benchmarking and Systems Performance

Cisco/Oracle Surgery on SPECjEnterprise

leave a comment »

Last month I had gum surgery. It’s funny how they market it to you. You only hear about how wonderful it will be for you. If you ever knew about some of the specifics of recovery, you might never do it. Ice packs, tea bags, and a lot of Motrin. A big bandage in your mouth and stitches that take weeks to come out. Sometimes you wonder if all of that was really worth it.

And that’s how I felt about Oracle’s new press release claiming a “world-record” Cisco/Oracle SPECjEnterprise2010 Java benchmark result. In order to get an astounding 3% or so lead, here’s what you need to know Cisco and Oracle had to do:

  • They had to use two application systems instead of one – two footprints, two servers with connections to manage.
  • They used only 4 JEE instances per server – how realistic is that?
  • They had to use a WebLogic version that isn’t even available until May.
  • They needed much more storage.
  • IBM is #1 in this benchmark for single node results, surpassing Oracle by 76%. (1)

  • One of the most common questions I got after the gum surgery is whether I lost any weight. Surely this must be a great way to diet. Alas, no — during recovery my two major food groups were pasta and tapioca pudding.

    ************************************************

    (1)SPECjEnterprise2010 result with WebSphere Application Server V7 on IBM Power 780 (64 cores, 8 chips) and DB2 9.7 on IBM Power 750 Express (32 cores, 4 chips) of 16,646.34 EjOPS. Oracle WebLogic Server 11g and Oracle Database 11g Release 2 on two Cisco Unified Computing System B440 MI Blade Servers, 17,301.86.44SPECjEnterprise2010 EjOPS. Oracle WebLogic Server Standard Edition 10.3.3 on Oracle SPARC T3-4 (4 chips, 64 cores app, 2 chips, 32 cores db) 9,456 SPECjEnterprise2010 EjOPS.
    Source: http://www.spec.org. Results current as of 3/17/11.

    SPEC, SPECint, SPECfp, SPECjbb, SPECweb, SPECjAppServer, SPECjEnterprise, SPECjvm, SPECvirt, SPECompM, SPECompL, SPECsfs, SPECpower, SPEC MPI and SPECpower_ssj are trademarks of the Standard Performance Evaluation Corporation (SPEC).

    The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.

    technorati tags: , , , , ,,,,,,,,,,,

    About these ads

    Written by benchmarkingblog

    March 17, 2011 at 11:56 am

    Leave a Reply

    Fill in your details below or click an icon to log in:

    WordPress.com Logo

    You are commenting using your WordPress.com account. Log Out / Change )

    Twitter picture

    You are commenting using your Twitter account. Log Out / Change )

    Facebook photo

    You are commenting using your Facebook account. Log Out / Change )

    Google+ photo

    You are commenting using your Google+ account. Log Out / Change )

    Connecting to %s

    %d bloggers like this: