October 2012 - mpiwg-rma - mailman3.cels.anl.gov

Re: [Mpi3-rma] FW: MPI_Accumulate
by Jeff Hammond 02 Oct '12

02 Oct '12

MPI-3 11.7.1 describes the atomicity of accumulate. Jeff On Tue, Oct 2, 2012 at 11:48 AM, Jeff Hammond <jhammond(a)alcf.anl.gov> wrote: > My understanding has always been that MPI_Accumulate was and is atomic > at granularity of the underlying built-in datatype, as opposed to > byte-wise atomic like MPI_Put. This is the whole reason that > MPI_Accumulate w/ MPI_REPLACE exists. Without the additional > atomicity guarantee, that feature would be pointless. > > Jeff > > On Tue, Oct 2, 2012 at 11:38 AM, Richard Graham <richardg(a)mellanox.com> wrote: >> resending >> >> >> >> From: Richard Graham >> Sent: Tuesday, October 02, 2012 12:22 PM >> To: MPI 3.0 Remote Memory Access working group >> (mpi3-rma(a)lists.mpi-forum.org) >> Subject: MPI_Accumulate >> >> >> >> What are the requirements placed on MPI_Accumulate if more than one mpi >> process tries to update the same location ? Does MPI provide any >> consistency promises, or is it up to the application to guarantee these ? I >> see that the get accumulate routines are defined to be atomic, but donï¿½t see >> the same requirement for accumulate. >> >> >> >> Thanks, >> >> Rich >> >> >> _______________________________________________ >> mpi3-rma mailing list >> mpi3-rma(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma > > > > -- > Jeff Hammond > Argonne Leadership Computing Facility > University of Chicago Computation Institute > jhammond(a)alcf.anl.gov / (630) 252-5381 > http://www.linkedin.com/in/jeffhammond > https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond -- Jeff Hammond Argonne Leadership Computing Facility University of Chicago Computation Institute jhammond(a)alcf.anl.gov / (630) 252-5381 http://www.linkedin.com/in/jeffhammond https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond

1 0

Re: [Mpi3-rma] FW: MPI_Accumulate
by Jeff Hammond 02 Oct '12

02 Oct '12

My understanding has always been that MPI_Accumulate was and is atomic at granularity of the underlying built-in datatype, as opposed to byte-wise atomic like MPI_Put. This is the whole reason that MPI_Accumulate w/ MPI_REPLACE exists. Without the additional atomicity guarantee, that feature would be pointless. Jeff On Tue, Oct 2, 2012 at 11:38 AM, Richard Graham <richardg(a)mellanox.com> wrote: > resending > > > > From: Richard Graham > Sent: Tuesday, October 02, 2012 12:22 PM > To: MPI 3.0 Remote Memory Access working group > (mpi3-rma(a)lists.mpi-forum.org) > Subject: MPI_Accumulate > > > > What are the requirements placed on MPI_Accumulate if more than one mpi > process tries to update the same location ? Does MPI provide any > consistency promises, or is it up to the application to guarantee these ? I > see that the get accumulate routines are defined to be atomic, but donï¿½t see > the same requirement for accumulate. > > > > Thanks, > > Rich > > > _______________________________________________ > mpi3-rma mailing list > mpi3-rma(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma -- Jeff Hammond Argonne Leadership Computing Facility University of Chicago Computation Institute jhammond(a)alcf.anl.gov / (630) 252-5381 http://www.linkedin.com/in/jeffhammond https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond

1 0

[Mpi3-rma] FW: MPI_Accumulate
by Richard Graham 02 Oct '12

02 Oct '12

resending From: Richard Graham Sent: Tuesday, October 02, 2012 12:22 PM To: MPI 3.0 Remote Memory Access working group (mpi3-rma(a)lists.mpi-forum.org) Subject: MPI_Accumulate What are the requirements placed on MPI_Accumulate if more than one mpi process tries to update the same location ? Does MPI provide any consistency promises, or is it up to the application to guarantee these ? I see that the get accumulate routines are defined to be atomic, but don't see the same requirement for accumulate. Thanks, Rich

1 0