[Mpi-forum] another request for iaccept

[William Gropp]

Some of the rationales do not apply.  In particular, I don’t believe these are important for Accept:

1) Low overhead (justifies Isend/Irecv etc.)
2) Scarcity of threads (e.g., the BlueGene/L rationale)

There are some interactions with multiple-competion routines and limitations in the generalized requests, but fixing generalized requests would be a more general solution.

Bill

William Gropp
Director, Parallel Computing Institute
Thomas M. Siebel Chair in Computer Science
Chief Scientist, NCSA
University of Illinois Urbana-Champaign





On Feb 25, 2016, at 2:40 PM, Jeff Squyres (jsquyres) <jsquyres at cisco.com> wrote:

>> On Feb 25, 2016, at 2:01 PM, William Gropp <wgropp at illinois.edu> wrote:
>> 
>> Part of that proposal must explain why using Accept in a separate thread (as suggested by this very post) isn’t sufficient.
> 
> Is it different than the rationale for other non-blocking MPI operations?
> 
> -- 
> Jeff Squyres
> jsquyres at cisco.com
> For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/
> 
> _______________________________________________
> mpi-forum mailing list
> mpi-forum at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum