Benchmarking and systems performance - hosted by Elisabeth Stahl

Elisabeth Stahl

Blog Authors:  ELISABETH STAHL   is Chief Technical Strategist and Executive IT Specialist, IBM Systems and Technology Group, and has been working in systems performance for over 25 years.

  • Add a Comment
  • Edit
  • More Actions v
  • Quarantine this Entry

Comments (2)

1 localhost commented Permalink

Elisabeth,<div>&nbsp;</div> You did not comment on an important aspect of this benchmark results, the J2EE application server results:<div>&nbsp;</div> The HP Integrity blade delivered a 22% better overall result, with 8% fewer cores which translates into a 32% better per core result for Itanium at 1.6 GHz than POWER6 delivered at 4.0 GHz. And the HP result is the absolute world record to date.<div>&nbsp;</div> Here are the facts:<div>&nbsp;</div> HP BL870c, 1.6GHz 24M Itanium, 48chips/96cores/2 core per chip, 12 Nodes, 17,046,4 JOPS@Standard. Which results in 177.6 JOPS/core<div>&nbsp;</div> IBM JS22, 4.0 GHz POWER6, 52chips/104cores/2 core per chip, 26 Nodes, 14,004.42 JOPS@Standard. Which results in 134.7 JOPS/core

2 localhost commented Permalink

JHannan, Thanks for your comments. I agree that it is important to look at the database and the application servers for this benchmark. I focus on the database server as an extremely important part of the configuration. 2.8 times the number of cores is also very different from an 8% difference !

Add a Comment Add a Comment