[Beowulf] SPEC CPU 2006 released

Keith D. Underwood kdunder at sandia.gov
Mon Sep 4 09:33:58 PDT 2006

> Hmmm.... we recently responded to a government RFP where they
> "require"
> runs on the hardware from the spec suite.
> [soap box]
>   This is counter productive IMO as the spec suite really doesn't do
> much for you in terms of meaningful performance measurements.  Does
> about the same as Linpack/HPL.  Yeah, some people argue otherwise, and
> compiler vendors and hardware OEMs/resellers work really hard to put
> out
> great benchmark data.  I don't believe such data is terribly
> meaningful,
> and customers who rely upon it for purchase decisions may wind up
> making
> decisions in part based upon "data" which has little intrinsic value
> to
> the tasks at hand.
> The most important benchmark is the one that uses the same code you
> use
> in the way you are going to use it.  Anything else is an entropy
> generator.
> [/soap box]

While I wholeheartedly agree that you should use the code you are going
to run as a benchmark, that is not always practical.  One thing SPEC
does nicely that Linpack/HPL does not is to make sure that it is a
"general purpose" platform.  I can imagine lots of machine
configurations that one might not want that would scream on Linpack/HPL
and would not run SPEC...


Keith D. Underwood                            Scalable Computing Systems
Senior Member of Technical Staff            Sandia National Laboratories
kdunder at sandia.gov

More information about the Beowulf mailing list