On May 24, 2012, at 23:48 , Dave Goodell wrote: > On May 24, 2012, at 10:34 PM CDT, George Bosilca wrote: > >> On May 24, 2012, at 23:18, Dave Goodell <good...@mcs.anl.gov> wrote: >> >>> So I take back my prior "right". Upon further inspection of the text and >>> the MPICH2 code I believe it to be true that the number of the elements in >>> the recvcounts array must be equal to the size of the LOCAL group. >> >> This is quite illogical, but it will not be the first time the standard is >> lacking some. So, if I understand you correctly, in the case of an >> intercommunicator a process doesn't know how much data it has to reduce, at >> least not until it receives the array of recvcounts from the remote group. >> Weird! > > No, it knows because of the restriction that $sum_i^n{recvcounts[i]}$ yields > the same sum in each group.
I should have read the entire paragraph of the standard … including the rationale. Indeed, the rationale describes exactly what you mentioned. Apparently the figure 12 on the following [MPI Forum blessed] link is supposed to clarify any potential misunderstanding regarding the reduce_scatter. Count how many elements are on each side of the intercommunicator ;) george. > The way it's implemented in MPICH2, and the way that makes this make a lot > more sense to me, is that you first do intercommunicator reductions to > temporary buffers on rank 0 in each group. Then rank 0 scatters within the > local group. The way I had been thinking about it was to do a local > reduction followed by an intercomm scatter, but that isn't what the standard > is saying, AFAICS.