[HPGMG Forum] Do we want the benchmark to go into intrinsics?

Sam Williams swwilliams at lbl.gov
Tue Apr 29 16:53:31 UTC 2014


Were any of the codes Bert/Ian analyzed manually simdized for QPX ?

I want to keep reference separate from optimized.  I didn't want to manually simdize anything until the community accepted HPGMG as a useful benchmark and agreed on the definition.  I also plan on keeping any of my optimizations private until published.



On Apr 29, 2014, at 9:43 AM, Jed Brown <jed at jedbrown.org> wrote:

> Sam Williams <swwilliams at lbl.gov> writes:
> 
>> I think there is a difference in saying 
>> - we need a high-quality implementation to tease out the kiviat characteristics in order to showcase the compute requirements of FEM
> 
> We need at least this.
> 
>> - we need a high-quality reference implementation that's been manually optimized for Mira.
> 
> So long as the performance counters reside on BG/Q, the first point
> implies some optimization for Mira.  I don't see having the best
> possible implementations in the public repository as a problem (indeed,
> I think it would be great).  So long as it does not cause portability
> problems, I think we are limited only by implementation effort.  I'm
> almost done with these tweaks.



More information about the HPGMG-Forum mailing list