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 (4)

1 localhost commented Permalink

I am a confused by the JS22 entries in the new IBM System p / Blade Center Performance Report (Nov 7, 2007). For the JS22 P6/4, the L2/L3 Cache (MB) is shown as "4/-". Shouldn't this be "16/-"? That is, elsewhere in the report, the convention seems to be that L2/L3 Cache is the total in the system, and I am assuming that the JS22's 4 cores each have 4MB L2 cache. I guess it could be that the JS22's core each have 1MB of L2 cache, but that contradicts other specs on your website. Thanks for clarifying.

2 localhost commented Permalink

Mark - Thanks so much for your comment. You are absolutely correct and I have submitted the update. Thanks !

3 localhost commented Trackback

Was Hyperthreading enabled on the DL380's in the test? Hyperthreading can slow the performance of CPU intensive calculations by as much as 40%

4 localhost commented Permalink

Mark - the SPEC report for the DL380 run can be found at http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02042.html. I don't see any mention of hyperthreading so I would recommend confirming with SPEC or HP. Thanks !

Add a Comment Add a Comment