[Mpi-forum] Proposal for the MPI-3 standardization process

[William Gropp]

The requirement for an implementation is understandable but is a  
departure from MPI 1 and MPI 2 - there we only required a convincing  
argument that the requirements were possible, not an example code.  I  
think this worked reasonably well (with some pluses and minus) and  
believe that the MPI 1 and MPI 2 rule should be sufficient.

Bill

On Dec 7, 2010, at 7:29 PM, Graham, Richard L. wrote:

> The attached document is a summary of our discussion yesterday - we  
> will talk about this tomorrow (Wed) as see if there are updates that  
> should be made.  Please come prepared with comments.
>
> Thanks,
> Rich
> <standardization_process.docx><ATT00001..txt>

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