[Beowulf] crunch per kilowatt: GPU vs. CPU

Craig Tierney Craig.Tierney at noaa.gov
Tue May 19 12:19:23 PDT 2009

Nifty Tom Mitchell wrote:
> On Mon, May 18, 2009 at 01:35:43PM -0400, Joe Landman wrote:
>> David Mathog wrote:
>>> Although the folks now using CUDA are likely most interested in crunch
>>> per unit time (time efficiency), perhaps some of you have measurements
>>> and can comment on the energy efficiency of GPU vs. CPU computing?  That
>>> is, which uses the fewest kilowatts per unit of computation.  My guess
>> Using theoretical rather than "actual" performance, unless you get the  
>> same code doing the same computation on both units:
> Another issue is the upgrade cost and upgrade potential.  Looking across
> my room I see an older AMD-64bit single core Athlon box.   It would be
> less expensive and much less of a task to swap out and upgrade the GFX
> card than it would be to update the processor.   In some cases multiple
> GFX cards are also possible.
> Also it makes little sense for me to upgrade the GFX card based on
> graphics display needs only.   If I toss CUDA in the mix a high end
> GFX card upgrade does get interesting.

It isn't that easy though.  First, does that system have a PCI-E slot,
or in particular a Gen2 slot?  Maybe changes to the PCI-E slots will
slow down so that you can efficiently use the GPU cards.  Second, does
that box have enough overhead in its power supply to drive an additional
150-250W needed for the GPU?  Does it have the right 6 or 8-pin connectors
necessary to power the card?  Does the full size GPU card fit in the box
you have?

High powered GPUs (and multiple ones in a box) can be lots of fun to deal


Craig Tierney (craig.tierney at noaa.gov)

More information about the Beowulf mailing list