[mvapich-discuss] truly one-sided communication

[sreeram potluri]

Dear Thiago,

I you can spare a thread, making it call MPI_iProbe on a dummy tag will help
progress in the background. From your description, we think that process 0
is the source and is computation intensive. In this case having an extra
thread at this process should not have a negative impact. However, you will
have to initialize MPI with MPI_THREAD_MULTIPLE.

MV2 uses process binding by default but in multi-threaded environments, it
is suggested to disable binding and let OS take care of it. Please see this
section of our user guide:

http://mvapich.cse.ohio-state.edu/support/user_guide_mvapich2-1.6rc2.html#x1-530006.12


Please let us know if this works and if you have any other questions.

Sreeram Potluri

On Wed, Jan 19, 2011 at 11:21 AM, Thiago <thiago at sci.utah.edu> wrote:

> Ah pity! I need this for a paper due in a few days.  Hmm, all I really need
> is an RMA read from the target memory which is read-only and never modified.
> I take it this means that synchronization is unneeded? Aside from waiting
> for MPI-3 to hopefully have this as an option (please!), how painful would
> it be to hack MVAPICH2 to simply throw out the synchronization so that when
> I call MPI_Get it only does a single RMA read over Infiniband and the
> lock/unlock skip the communication?  If that's too complicated, what would
> be the next best option?  I have 64 nodes, each running 8 threads where each
> thread does computation and random MPI_Get communication.  Is there
> something, (perhaps unportable and only works on 1.6-rc2!) that I could do
> that would let me call MPI_Get without having to have the target know this
> is happening?  Perhaps using MPI_MODE_NOCHECK and doing "undefined" behavior
> that actually ends up working just this time?  If I must do a standard
> passive communication, what's the best way to ensure my performance doesn't
> tank?  Is there a cheap MPI function I can have a thread do in a while loop
> to ensure that progress is made?
>
> Thanks,
> Thiago
>
>
>
> On 1/19/2011 7:22 AM, sreeram potluri wrote:
>
>> Dear Thiago,
>>
>> Thanks for your email. Currently, MVAPICH2 supports truly one-sided
>> communication for the active modes of synchronization (Fence and
>> Post-Wait/Start-Complete) available in MPI-2 RMA. The truly one-sided
>> designs for passive mode are in the pipeline for the next (1.7) release of
>> MVAPICH2.
>>
>> Thank you
>> Sreeram Potluri
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20110119/22eb09a9/attachment.html