[Mpi-forum] ABI

[Jeff Squyres (jsquyres)]

I think there was one other key takeaway from the ABI WG: it would seem like a good public solution would be to develop an open source morph MPI layer that can talk to any MPI implementation on the back-end.

A few people have tried this over the years with moderate success, but I don't think that anyone has fully implemented MPI3 and pushed it to the entire HPC community.  So none of them have gotten too widely adopted.

My $0.02 is that a public, free, open source Morph MPI layer with an active community behind it would still be a very good starting point for whatever could be achieved via some kind of binary compatibility.


On Dec 5, 2013, at 12:42 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:

> "Supalov, Alexander" <alexander.supalov at intel.com> writes:
>> I am afraid I cannot comment on particular snippets of other people's
>> email 5-6 year after the events. If you wish, we can get this trail
>> offline and go through the material with you.
> 
> No worries.  I'm happy to discuss further, though this list probably
> isn't the best place.  Are the interested parties subscribed to
> discuss at mpich.org?
> _______________________________________________
> mpi-forum mailing list
> mpi-forum at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum


-- 
Jeff Squyres
jsquyres at cisco.com
For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/