[Mpi3-rma] Updated Proposal 1
Barrett, Brian W
bwbarre at sandia.gov
Mon Nov 29 10:06:40 CST 2010
On Nov 26, 2010, at 3:47 PM, Torsten Hoefler wrote:
>>> 2) Keith and Brian, could you please elaborate on the arguments against
>>> allowing multiple elements (count>1) in MPI_Get_accumulate? I remember
>>> there was some discussion about buffering and failures if one wanted to
>>> support it in hardware but I don't remember what the issues were. It
>>> seems like one could simply pipeline the hardware operations or just
>>> fall back to a software implementation if count is bigger than a certain
>> With pipelining, I assume atomicity is only per element? I do not see
>> how you could realize anything else in hardware. I don't see any
>> value in multi-element RMW if it is element-wise atomic. One could
>> just as easily send multiple messages.
> Correct, the current RMA doesn't offer more than element-wise atomicity
> and we don't plan to extend this. You are right, the only benefit would
> lie in reduced overheads which isn't all that much. I'm not sure if
> full-message atomicity would be easy/fast to implement.
I think we could do multi-element with the very loose element-wise guarantee. At one point, there was discussion about strengthening that guarantee, which would make the multi-element Get_accumulate painful.
Brian W. Barrett
Dept. 1423: Scalable System Software
Sandia National Laboratories
More information about the mpiwg-rma