Boost logo

Boost-MPI :

Subject: Re: [Boost-mpi] Point 2 point free collectives.
From: MM (finjulhich_at_[hidden])
Date: 2016-10-25 16:18:05


On 25 October 2016 at 10:35, alainm <alain.miniussi_at_oca.eu> wrote:

> scatterv and gatherv are now point to point free in globopt branch. They
> share most of their implementation with their non variable counterpart as
> far as non MPI type are concerned:
> Since the size of a non MPI type does not have a fixed size footprint
> anyway, we typically need to serialize, communicate the archives size ans
> send them as byte arrays.
>
> All_gatherv implementation is still missing.
>
> Cheers
>
> On 20/10/2016 19:24, alainm wrote:
>
>> I just did gather in the same spirit, the slight difference is that the
>> size of the archive are not homogeneous.
>> So, although it is called gather, it is implemented through MPI_Gatherv.
>>
>> Which raises another issue, it is likely that the same interface could
>> serve the "v" version of the function, thus simplifying the API. We only
>> need to communicate the numbers of array slots along with the size of the
>> archives.
>>
>> Cheers
>>
>>
>> On 20/10/2016 09:17, Lorenz Hübschle-Schneider wrote:
>>
>>> Hi everyone,
>>>
>>> On 19/10/16 10:00, alainm wrote:
>>>
>>>> b) Not so trivial calls need to go though serialization, and manipulate
>>>> the
>>>> archive using point to point communication
>>>>
>>>
>>> I don't understand why archives would need to be communicated using
>>> point-to-point communication. I've posted this before with no response, but
>>> under which circumstances would something like
>>>
>>>> https://gist.github.com/lorenzhs/79dab54552fd1f9381da
>>>>
>>> fail?
>>>
>>> Cheers,
>>> Lorenz
>>> _______________________________________________
>>> Boost-mpi mailing list
>>> Boost-mpi_at_[hidden]
>>> http://lists.boost.org/mailman/listinfo.cgi/boost-mpi
>>>
>>
>>
> --
> Alain Miniussi
> Pole Génie Logiciel Scientifique
> Observatoire de la Côte d'Azur
> Blv de l'Observatoire, Nice
>
> _______________________________________________
> Boost-mpi mailing list
> Boost-mpi_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/boost-mpi
>

I'm sorry I didn't follow the discussion, but how are the scatterv and
gatherv you're proposing in your branch different from the ones already in
latest boost ?



Boost-Commit list run by troyer at boostpro.com