(j3.2006) [MPI3 Fortran] Feedback from Fortran J3 meeting
Thu May 29 13:33:37 EDT 2008
On May 29, 2008, at 12:54 PM, Aleksandar Donev wrote:
> I believe that it is a mistake to not
> spend the time making a general facility. There are a lot more
> libraries out
> there than MPI.
I agree. Indeed , it may be easier to write support
for general asynchronous routines than to try to specialize
for MPI. Other communications and storage libraries
should not be prejudiced.
The shelf life of some MPI applications may be shorter
than we estimate from here. Most MPI codes known to me
are under active development to incorporate new science,
and the advantages of coarrays for such modifications are great.
More information about the J3