Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

Subject: Re: [OMPI users] Optimal 3-D Cartesian processor mapping
From: Tom Rosmond (rosmond_at_[hidden])
Date: 2012-04-24 15:33:54


Jeff,

Yes, I would be interested in such a plugin. But be advised that I am
strictly a fortran programmer, so if it requires any C/C++ talent, I
would be in trouble. So maybe, before jumping into that, I would like
to be able to look at what processor/node mapping Open-mpi is actually
giving me. Is there an environmental variable or an MCA option I can
add to my 'mpirun' command line that would give that to me? I am
running 1.5.4.

T. Rosmond

On Tue, 2012-04-24 at 15:11 -0400, Jeffrey Squyres wrote:
> On Apr 24, 2012, at 3:01 PM, Tom Rosmond wrote:
>
> > My question is this: If the cartesian mapping is done so the two
> > spacial dimensions are the 'most rapidly varying' in equivalent 1-D
> > processor mapping, will Open-mpi automatically assign those 2 dimensions
> > 'on-node', and assign the 'ensemble' dimension as the slowest varying
> > and across nodes? If not, how can we guarantee this to happen?
>
>
> Sadly, the MPI mapping functions in Open MPI are effectively no-ops -- if you ask for re-ordering, OMPI says "sure!" but then gives you back exactly the same mapping. :-(
>
> That being said, the mapping functionality is actually a plugin in Open MPI (in the "topo" framework). It would not be too difficult to write a plugin for the mapping that you want.
>
> Would this be of interest to you? I'd be happy to talk you through the process of writing such a plugin.
>