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

Funny how IBM needed to run a clustered p575 system, (which I doubt many customers would choose to run a high end DB) to beat an almost 2-year old Sun E25K 1.5GHz system that was NOT clustered for the TPC-H 10TB benchmark. <div>&nbsp;</div> Why didn't IBM run its high end p595 system on TPC-H 10-TB or 30-TB like it did on TPCH 3-TB, where the Sun E25K beat the p595 by 14% on performance AND 31% better price/performance (includes licensing costs)! <div>&nbsp;</div> At least HP is showing initiative by publishing the same system across multiple DB sizes to show scaling! And where are the p570 Power6 TPC-H scores? Can't do well on large DB's? And don't tell me the 14-year old TPC-C is a good benchmark to showcase DB performance!

2 localhost commented Permalink

Phil - Thanks for your comments. Note that IBM did not "need" to run a clustered system. IBM publishes both clustered and nonclustered results on benchmarks to highlight different systems and configurations. On your 3TB comparison, make sure that you note that the Sun E25K was 144 cores while the p5-595 was only 64 cores. And as I've said before, be careful what you wish for when asking about POWER6 TPC-H results.

Add a Comment Add a Comment