When I hear "T3" I can't help but think of a wonderful program I've been a part of, where the top notch experts in some area are taught about something new and then go out and teach others. Train the Trainer, it's called.
T3 is also the name for Oracle's new SPARC T3 processor and systems. If only the new T3 benchmarks were so top notch. Let's take a closer look:
Oracle touts a T3 SPECjEnterprise2010 result. A SPECjEnterprise2010 IBM POWER7 result with WebSphere and DB2 was actually 51% greater than the performance per core of this new Oracle SPARC T3 server result.(1) And IBM is #1 in this benchmark with an X5 system, again using DB2 and WebSphere.(2) Oracle touts a T3 SPECjvm2008 result as best performance in the industry. Too bad there are only old Sun and Apple results to compare with. Why is this benchmark so unpopular? Most of the new Oracle SPARC T3 benchmarks announced were actually Oracle's own applications -- essentially internal tests. Even something called an "in-house workload" was used. Oracle highlights an increase over performance of their previous generation systems. And all I want to say is, I sure hope so. Or why even bother?
(1)SPECjEnterprise2010 Oracle WebLogic Server 11g on a Oracle SPARC T3-4 server (4 chips, 64-core) and Oracle Database 11g on SPARCT3-2 server (2 chips, 32-core) , 9,456.28 SPECjEnterprise2010 EjOPS vs. WebSphere Application Server V7 on IBM Power 750 Express (4 chips, 32-core) and DB2 9.7 on IBM BladeCenter PS702 Express (2 chips, 16-core), 7,172.93 SPECjEnterprise2010 EjOPS.
(2)SPECjEnterprise2010 WebSphere Application Server V7 on IBM System x HS22 (12 chips, 72-core) and DB2 9.7 on IBM System x3850 X5 (4 chips, 32-core), 15,829.86 SPECjEnterprise2010 EjOPS.
Source: www.spec.org. Results current as of 9/29/10.
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: IBM, systems, performance, SPECjvm2008, SPECjEnterprise2010,750,SPARC,T3,WebSphere,PS702,benchmark,Oracle,OpenWorld,X5,Sun,Java,Apple,POWER7,x3850,DB2,HS22,Weblogic,benchmarking