[Mpi-forum] MPI One-Sided Communication

[William Gropp]

Thanks, Greg.

One other thing that I'd like to remind everyone is that there's a  
difference between "MPI (the standard)" and "the performance of a  
particular MPI implementation".  These are all too often confused, and  
there are a lot of suboptimal MPI implementations of the MPI RMA  
routines out there.  In particular, in most cases, if the asymptotic  
bandwidth of MPI (whether it is send/recv or put) isn't as high as  
some other approach, that almost always means that the MPI  
implementation is doing something inefficient, not that "MPI" is  
inefficient.  There was one vendor implementation, for example, that  
made unnecessary buffer copies - that's not the fault of MPI, just an  
unoptimized implementation.

Bill

On Apr 24, 2009, at 12:01 PM, Greg Lindahl wrote:

> On Fri, Apr 24, 2009 at 07:38:39AM -0500, Jeff Hammond wrote:
>
>> You're comparing ("scaled better than") apples ("an implementation on
>> top of 2-sided MPI-1 send/recv") and oranges ("real one-sided
>> hardware").  Please post the whitepaper (I cannot find it online) and
>> clarify what you mean here.
>
> QLogic seems to have hidden it here:
>
> http://www.qlogic.com/SiteCollectionDocuments/Education_and_Resource/whitepapers/whitepaper1/RealApplications-paper.pdf
>
> I used the Siosi6 benchmark, since I needed something that had public
> benchmark numbers available.
>
> Our implementation used 1/4 of the cores as "memory servers" which
> held the global memory and served as the counterparty to one-sided
> operations. Despite wasting 25% of the cores, it scaled great.
>
> Now this result won't apply to all one-sided programs; GA's model and
> usage is somewhat different from most other one-sided models.
>
> -- greg
>
>
> _______________________________________________
> mpi-forum mailing list
> mpi-forum at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum

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