[Mpi3-rma] Alternative RMA discussion
William Gropp
wgropp at illinois.edu
Sun Dec 13 12:30:54 CST 2009
I think part of the way to address this (too many sync functions) is
follow through on another of Torsten's recommendations - provide
better user-oriented material on using the interface. Most libraries
these days have large numbers of methods/functions; this complaint
usually means "I can't figure out which routines to use" not "there
are too many routines".
Bill
On Dec 13, 2009, at 11:48 AM, Torsten Hoefler wrote:
>>
>> Since one of the criticisms of the current interface is too many
>> synchronization functions, it may be worth looking into whether the
>> synchronization requirements can be relaxed in some way in the cache
>> coherent case.
> I would like to hear more about this criticism. I think the current
> set
> of functions is rather elegant (only the group argument in the PSCW
> model is a bit uncommon). But fence reflects BSP nicely and lock/
> unlock
> (if we have p2p windows) is a nice remote update mechanism.
William Gropp
Deputy Director for Research
Institute for Advanced Computing Applications and Technologies
Paul and Cynthia Saylor Professor of Computer Science
University of Illinois Urbana-Champaign
More information about the mpiwg-rma
mailing list