I heard this morning that at OpenWorld they have beautiful white tents lining the streets. And gourmet coffee. And candy buffets.
No wonder everyone wants to get into this act. Play in the Lego playpens, where you get beer and the Black Eyed Peas.
Cisco has joined the party by claiming a SPECjAppServer2004 record, announced by Oracle at this event. Let's look at this result a little more closely.
SPECjAppServer2004 is old news. In fact, IBM has not published on this benchmark for almost two years. And why is that? Because the SPECjEnterprise2010 benchmark measures full-system performance for Java Enterprise Edition application servers, databases and supporting infrastructure and expands the scope of the SPECjAppServer2004 benchmark. Even Oracle knows this since their new SPARC T3 result used SPECjEnterprise2010. Remember, it's 2010, not 2004. The IBM result that was published almost two years ago on a POWER5 system with WebSphere and DB2 had better database performance per core than this brand new Cisco result running Oracle software.(1) With this result, Cisco is claiming a dual node record. Of the 95 SPECjAppServer2004 results published, only 9 were dual node. 6 of those were Oracle/Sun. And the result that Cisco is touting that they beat? That's a result that Oracle published.
Cisco may think that Oracle is their friend. But Cisco better be careful that nothing is put in their coffee for that one.
(1) IBM SPECjAppServer2004 result of 22,634.13 JOPS using WebSphere Application Server V188.8.131.52 on IBM BladeCenter HS21 blades. Sixteen HS21 application servers running Red Hat Enterprise Linux 5 Server Update 2 were used. Each app server consisted of two quad-core Intel Xeon X5470 3.325 GHz processor chips (8 cores, 2 chips, 4 cores/chip, 16384 MB memory). The database server was an IBM System p5-595 configured with dual-core 2.1 GHz processor chips (64 cores, 32 chips, 2 cores/chip, 393216 MB memory) running IBM DB2 9.5 on IBM AIX 5L v5.3. DS4800 storage controllers were used. Oracle Web Logic Server running on two Cisco B230 M1 Blade Servers each with two, eight-core Intel Xeon X7560 2.26 GHz processors, database - 4 8-core (32-core) Cisco B440, 11,283.80 SPECjAppServer2004 JOPS@Standard.
Source: www.spec.org. Results current as of 9/22/10.
SPEC, SPECint, SPECfp, SPECjbb, SPECweb, SPECjAppServer, SPECjEnterprise, 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, SPECjAppServer2004, benchmark,Oracle,OpenWorld,Exalogic,cloud,Java,HP,POWER,DL980,DB2,Exadata,Weblogic,benchmarking