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 Trackback

That is a totally specious argument. While valid regarding subjective ratings like you gave in your examples, it is completely invalid regarding objective ratings. Sure, having audited, third party tests would be great, but failing that, there is no reason to doubt self-measured objective ratings. For instance, if someone says "I am a handsome man", I might take the statement with a grain of salt. But if he says "My car is getting 27 miles to the gallon", I would be inclined to believe him.<div>&nbsp;</div> The fact remains that the data Sun is providing self-measured, IBM does not provide at all. As suspect as the self-measured data is, how much more suspect is data not provided? While it is disappointing that Sun has not published full specpower numbers, as has already been pointed out by others, watts consumed while running other benchmarks is probably more useful to consumers anyway.<div>&nbsp;</div> In the interests of full disclosure, I am a Sun employee, but it is not my job to answer these kinds of posts. I just felt strongly about the injustice presented here.

2 localhost commented Permalink

How can data not provided be suspect? An industry benchmark has a disclosure report which vendors publish after conducting their benchmark(s), Sun by adding a new field and then stating no one else publishes this information and questioning their motives is also an injustice.<div>&nbsp;</div> Since Sun has not published SPECpower it is dismissed as maybe not being that important as other benchmarks. (Remimds me that TPC-C is also not considered relevant by Sun)<div>&nbsp;</div> But if some data is better than no data, maybe Sun should really start conducting and publishing TPC-C &amp; SPECpower benchmarks; TPC-C is a very useful benchmark that lots of consumers refer to. (I have heard the excuses but I have also seen Sun provide estimations when threatened with disqualification in tenders).

Add a Comment Add a Comment