> My experience at the MPI Forum meetings has been exactly the reverse of the
> above. In general, it is the vendors and implementors who have been arguing in
> favor of simplicity, and the users who have been saying "but I really need this
> fancy whiz-bang function".
I see - the balance has obviously shifted since MPI-1 days. It sounds like
this is a question of how high or low-level you are talking. I was talking
about a relatively low level. Perhaps I should have made it clear I was
speaking about the philosopy behind MPI-1. However, as Marc Snir pointed out
in his very clear perspective of the situation:
> and that this design is safer. The decisions may have been wrong, and
> CDS1 may be a superior design. But this is clearly not a fruitful debate
> for the MPI forum. Clearly, the MPI forum has to design MPI2 using the
> same design principles as MPI1, while other groups or individuals may
> push designs built on totally different principles.
Since I am no longer actively involved in the forum, I will now resume my
former silence, and continue to track developments as best I can from what
arrives in the mail :-)
Ian
-- Ian Glendinning European Centre for Parallel Computing at Vienna (VCPC) ian@vcpc.univie.ac.at Liechtensteinstr. 22, A-1090 Vienna, Austria Tel: +43 1 310 939612 WWW: http://www.vcpc.univie.ac.at/~ian/