Subject: Re: [Boost-mpi] non blocking collectives
From: alainm (alain.miniussi_at_[hidden])
Date: 2016-10-19 04:00:52
Just a few thoughts.
Boost.MPI mapping of MPI could be splitted in two categories, especially
a) the "trivial" calls (the one with MPI dataypes), on simple type, are
directly mapped to it MPI API counter part.
b) Not so trivial calls need to go though serialization, and manipulate
the archive using point to point communication (sometime, the size need
to be communicated first, then the data).
Adapting that to non blocking means delegating to MPI_IBcast for a) and
to non blocking point to point for b. Now, for case b, the surrent
implementation already uses non blocking point to point, but wait for
the completion of all requests before returning. I guess the async
version will need to offer a kind of request compound object that offer
a single API to handle both cases.
On 19/10/2016 08:10, MM wrote:
> On 17 October 2016 at 14:48, alainm <alain.miniussi_at_oca.eu> wrote:
>> MPI3 is not covered yet. It will probably be done at some points.
> I'll have a go at implementing ibroadcast based off the broadcast()
> Boost-mpi mailing list
-- Alain Miniussi Pole Génie Logiciel Scientifique Observatoire de la Côte d'Azur Blv de l'Observatoire, Nice
Boost-Commit list run by troyer at boostpro.com