[HPGMG Forum] HPGMG release v0.1

Jed Brown jed at jedbrown.org
Mon Jun 9 14:08:45 UTC 2014


Sam Williams <swwilliams at lbl.gov> writes:

> Are you referring to the time-to-solution figure or the DOF/s figure?

DOF/s, but both are relevant.

> Currently, the x-axis is synonymous with problem size.  i.e 128^3,
> 256^3, 384^4, ... 5120^3.  Thus you can comment on the the differences
> in time-to-solution for the same problem size on differing processor
> architectures with the caveat that there are 128^3 DOF per numa node.
>
> If you change this, then at a given x-coordinate, it would be showing time-to-solution for
>  128x128x128 on Mira / K
>  128x128x256 on Edison/Peregrine
>  128x256x256 on Hopper
> I'm not sure this helps.

The choice of problem size is arbitrary and normalizing per NUMA node is
friendly toward Hopper's network (bigger local sizes mean less network
stress).  It would be nice to add the extra dimension so that we could
see dynamic range alongside this performance.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 818 bytes
Desc: not available
URL: <https://hpgmg.org/lists/archives/hpgmg-forum/attachments/20140609/6b467312/attachment.bin>


More information about the HPGMG-Forum mailing list