[petsc-users] CPU vs GPU for PETSc applications

[Matthew Knepley]

On Thu, Mar 10, 2016 at 4:48 PM, Justin Chang <jychang48 at gmail.com> wrote:

> Matt,
>
> So what's an example of "doing a bunch of iterations to make sending the
> initial datadown worth it"? Is there a correlation between that and
> arithmetic intensity, where an application is likely to be more
> compute-bound and memory-bandwidth bound?
>

Say

  1) Send rhs and assume 0 initial guess

  2) Perform k matvecs for some Krylov solver

  3) Get back output vector

the elephant in the room here is that I have not specified a
preconditioner. It is here that GPUs have the most
trouble, but if you can find a GPU PC that works, and you use a bunch of
iterations before communicating back,
you can realize the 2x speed benefit over a modern CPU.

  Thanks,

     Matt


> Thanks,
> Justin
>
> On Thu, Mar 10, 2016 at 2:50 PM, Matthew Knepley <knepley at gmail.com>
> wrote:
>
>> On Thu, Mar 10, 2016 at 12:29 PM, Justin Chang <jychang48 at gmail.com>
>> wrote:
>>
>>> Hi all,
>>>
>>> When would I ever use GPU computing for a finite element simulation
>>> where the limiting factor of performance is the memory bandwidth bound? Say
>>> I want to run problems similar to SNES ex12 and 62. I understand that there
>>> is an additional bandwidth associated with offloading data from the CPU to
>>> GPU but is there more to it? I recall reading through some email threads
>>> about GPU's potentially giving you a speed up of 3x that on a CPU but the
>>> gain in performance may not be worth the increase in time moving data
>>> around.
>>
>>
>> The main use case is if you are being forced to use a machine which has
>> GPUs. Then you can indeed get some benefit
>> from the larger bandwidth. You need a problem where you are doing a bunch
>> of iterations to make sending the initial data
>> down worth it.
>>
>> It would certainly be better if you are computing the action of your
>> operator directly on the GPU, but that is much more
>> disruptive to the code right now.
>>
>>   Matt
>>
>>
>>> Thanks,
>>> Justin
>>>
>> --
>> What most experimenters take for granted before they begin their
>> experiments is infinitely more interesting than any results to which their
>> experiments lead.
>> -- Norbert Wiener
>>
>
>


-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20160311/bfef6b29/attachment.html>