Astaire and Rogers. Hot dogs and mustard. Yin and Yang. What would one be without the other?
When you look at the SPECjAppServer2004 benchmark it's imperative to realize that both the application server and the database server are integral to the performance of the system.
In fact, some would say that the database server is what you really need to look at here, even though many of us naively focus on the application server.
Which is exactly what Oracle did in their latest press release. If you look at the IBM database server that Oracle compares to, the IBM result per core was actually over 3 times the performance of Oracle's latest result.(1) Oh, and that IBM server. It wasn't a POWER7 system. It wasn't a POWER6+ or POWER6 system. It was a POWER5+ system from 2008.
One more thing to add. Oracle's other claims on Java benchmarks in the press release. They are from retired benchmarks that are as old as the hills.
App Server and Database. Just remember. Would you want your fish and chips without the chips?
(1) IBM BladeCenter HS21 cluster (3.3 GHz, 32 chips, 128 (16x8) cores) application server and IBM p5-595 64-core database server, 22,634.13 SPECjAppServer2004 JOPS@Standard, WebSphere Application Server V7, DB2 9.5 vs. Oracle WebLogic Server 11g on five-node Sun SPARC Enterprise T5440 cluster each with four 8-core UltraSPARC T2 Plus 1.6 GHz processors, 28,648.74 SPECjAppServer2004 JOPS@ Standard. Source: www.spec.org, Results current as of 6/15/10.
SPEC, SPECint, SPECfp, SPECjbb, SPECweb, SPECjAppServer, SPECjEnterprise, SPECompM, SPECompL, SPECsfs, SPECpower, SPEC MPI and SPECpower_ssj are trademarks of the Standard Performance Evaluation Corporation (SPEC).
Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.
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: IBM, systems, performance,