[Beowulf] Performance characterising a HPC application

Mark Hahn hahn at mcmaster.ca
Wed Mar 21 04:50:00 PDT 2007


>> http://www.sharcnet.ca/~hahn/m-g.C is a benchmark I'm working on.  it's
>> mainly set up to just probe bw and latency for every pair of nodes in a
>> cluster (obviously diagnostic).
>
> ... which is what the nxnlatbw microbenchmark distributed with MPICH
> does.

<shrug>.  I have no qualms about reinventing wheels.  but I also can't 
find this code - it doesn't seem to be in a copy of mpich2-1.0.5p3
I have sitting around, and google doesn't know the string.

in truth, I had more of microway's link-checker in mind.

> Perhaps you have other things in mind. But, how about a
> microbenchmark that does what actual programs do?

I would call that "the application".  seriously, a benchmark which 
does what actual apps do, but is somehow not the app?  is there some
reason to believe that there is not some sort of basis set of primitives
which actual app performance can be factored into?

regards, mark hahn.



More information about the Beowulf mailing list