[Mpi3-rma] [EXTERNAL] Re: MPI-3 UNIFIED model clarification
balaji at mcs.anl.gov
Sun Aug 4 18:18:33 CDT 2013
On 08/04/2013 06:15 PM, Jed Brown wrote:
> Pavan Balaji <balaji at mcs.anl.gov> writes:
>> SEPARATE is too expensive for such cases.
> What is the lightest weight way for the source to create an ordering at
> the target (which will be using MPI_Win_sync)? How does the overhead of
> that compare to MPI_Win_sync?
This is not an ordering issue if I'm talking about nonoverlapping
memory. If P0 is doing PUTs to P1s memory, and P1 is reading some other
nonoverlapping region of it's own memory, P0 and P1 need to synchronize
in some way if they use the SEPARATE model (e.g., by using EXCLUSIVE
locks). If they used the UNIFIED model, they can do these accesses
concurrently without any synchronization.
More information about the mpiwg-rma