Jump to content United States-English
HP.com Home Products and Services Support and Drivers Solutions How to Buy
»  Contact HP
HP.com Home

The Real Story about Real World Performance


Large Enterprise Business

» Products
» Business & IT Services
» Solutions
» Technologies
» Partners
» Support & Drivers
» Business Technology
» Media Center & Library
» Support & Troubleshooting
» Software & Drivers
Content starts here
text imageJanuary 2009: The SPECjappServer2004 (Java Application Server) benchmark is a modern benchmark for measuring the performance of Java 2 Enterprise Edition (J2EE) application servers. This benchmark heavily exercises all parts of the underlying infrastructure that make up the application environment, including server hardware, Java Virtual Machine software, database software, the network and more.

» The Real Story home page
» Real Story about POWER6
» Blog: Mission Critical Musings

IBM has written that the SPECjappServer2004 workload represents a, “…real-world customer environment.”2 On this point we both agree.

Did you know that the HP Integrity BL870c server blade significantly outperforms the POWER6 based IBM JS22 blade on this real world benchmark?

Please consider the following fact:

Fact 1: Using a configuration that had the same number of processor cores on the J2EE appserver systems, the HP Integrity BL870c blade server delivered greater than 50% more performance than the IBM JS22 POWER6 blade.

System Description SPECjappServer2004 JOPS@Standard App/Database/OS
HP Integrity BL870c server blade dual-core Intel® Itanium® 9100series 1.6GHz, 13 blades/52 chips/104 cores. Each blade: 8 cores/4 chips/ 2 cores per chip
Oracle WebLogic Server Standard Edition Release 10.3 HP–UX 11iv3 September 2008
IBM BladeCenter JS22 Express dual-core POWER 6 4.0GHz, 26 blades/52 chips/104 cores. Each blade: 4 core/2 chips/ 2 cores per chip
WebSphere 6.1 Application Server with EJB3 Feature Pack/DB2/AIX 5L V5.3
Results as of 01/12/2009. See: www.spec.org You'll be leaving HP.com
SPECjappServer2004is a trademark of the Standard Performance Evaluation Corp. (SPEC).
To learn more read: HP Integrity BL870c server performance brief

And the Intel Itanium beat IBM POWER6 even at a much lower processor clock frequency. Processor clock frequency alone is a poor measure of solution performance. As evidenced by this result HP provides real solution performance.

For more see www.hp.com/go/hpengineers. Video benchmark details 3
Better infrastructure performance means better business performance.

Balanced Performance

No one industry standard benchmark can be used entirely to forecast an enterprise’s unique set of performance needs. That is why we designed HP Integrity servers to have balanced design able to handle the varied workloads found in today’s real world.

See HP Integrity server performance and benchmarks to see how HP Integrity Servers deliver balanced performance across a wide range of workloads.

To learn more about HP Integrity servers see: www.hp.com/go/integrity

SPEC and the benchmark name SPECjAppServer2004 are registered trademarks of the Standard Performance Evaluation Corporation. Competitive benchmark results stated above reflect results published on www.spec.org You'll be leaving HP.com as of Jan 12, 2009. The comparison presented above is based on multi-node results using matching number of cores for the J2EE Appserver systems, from HP and IBM. For the latest SPECjAppServer2004 benchmark results, visit http://www.spec.org/jAppServer2004/ You'll be leaving HP.com.

1 See: http://www.spec.org/jAppServer2004/ You'll be leaving HP.com

2 IBM Press Release, Feb 15, 2008: http://www-03.ibm.com/press/us/en/pressrelease/23523.wss You'll be leaving HP.com,

“SPECjAppServer2004 benchmark, an independent, industry-standard benchmark that measures the scalability and performance of Java 2 Enterprise Edition (J2EE) application servers.

"This benchmark delivers an unprecedented, high level of performance in a real-world customer environment,"


http://www.ibm.com/developerworks/data/library/techarticle/dm-0606lau2/index.html You'll be leaving HP.com, “SPECjAppServer 2004 benchmark is the leader in J2EE benchmark. The topology diagram in Figure 1 You'll be leaving HP.com below illustrates a sample SPECjAppServer benchmark configuration. The SPECjAppServer2004 benchmark reflects the rigors of complex e-business applications and high transaction rate that are typically deployed in today's enterprise customer environments.”

3 Video Benchmark Details:
HP BL860c, 1.66GHz Itanium, 24chips/48cores/2 core per chip, 12 Nodes, 10,519.4 JOPS@Standard. 219.2 JOPS/core
IBM JS22, 4.0 GHz POWER6, 52chips/104cores/2 core per chip, 26 Nodes, 14,004.42 JOPS@Standard. 134.7 JOPS/core
Results as of 01/12/09, see: www.spec.org You'll be leaving HP.com SPECjappServer2004, is a trademarks of the Standard Performance Evaluation Corp. (SPEC).
Printable version
Privacy statement Using this site means you accept its terms
© 2011 Hewlett-Packard Development Company, L.P.