[petsc-users] Building with MKL 10.3

[Matthew Knepley]

On Mon, Mar 14, 2011 at 7:48 PM, Lisandro Dalcin <dalcinl at gmail.com> wrote:

> On 14 March 2011 20:48, Barry Smith <bsmith at mcs.anl.gov> wrote:
> >
> >  Eric,
> >
> >   With the current PETSc design that uses one MPI process per core I am
> not sure that it makes sense to use multi-threaded MKL since the processes
> are already using all the cores. I could be wrong.
> >
> >   Barry
> >
>
> Well, I think that PETSc should support sequential codes using
> multi-threaded blas-lapack and perhaps some OpenMP. Moreover, perhaps
> we could get some little speedup in MatMult for these kinds of
> users/applications.
>

I have never ever seen convincing evidence of this. First, you would need
enough bandwidth to satisfy
2+ cores. This is almost never the case. But suppose you do have this. Then
you would need a convincing
reason to use threads instead of MPI processes, which would mean data reuse.
But there is very little
reue here; it is mostly streaming.

   Matt


> --
> Lisandro Dalcin
> ---------------
> CIMEC (INTEC/CONICET-UNL)
> Predio CONICET-Santa Fe
> Colectora RN 168 Km 472, Paraje El Pozo
> 3000 Santa Fe, Argentina
> Tel: +54-342-4511594 (ext 1011)
> Tel/Fax: +54-342-4511169
>



-- 
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/20110314/5ccafaad/attachment.htm>