[Beowulf] Re: Re: Home beowulf - NIC latencies
jcownie at etnus.com
Fri Feb 18 06:38:23 PST 2005
> In a conversation with MPI and tool developers and I once mentioned
> that not defining a standard/mandatory mpi.h was probably a missed
> opportunity for improving interoperability of MPI. I was then told by
> a member of the MPI-1 Forum that this was done on purpose. This makes
> me think that we will not see an ABI definition for MPI any time soon.
I think this is to misunderstand the process.
The whole MPI process was informal. No-one gave the committee any power
to create a standard, it happened because people wanted it to happen and
were prepared to use the result.
MPI followed the format and style of HPF, and can be thought of as an
Open Source standard.
It was created as a result of user demand by people who were prepared to
put in the effort to do so, and was adopted because it met a need.
If an ABI for MPI is so important to you and of such value to your (and
Patrick's) clients, then there's nothing to stop you from formulating
such a standard, or, at least starting a project to create one.
If you're right about its importance, then all the MPI implementors will
follow your lead. If you're wrong, well, you wasted your time.
The point here is that doing this can be an informal process which
doesn't require "The MPI Forum" (whatever that is now !?) to endorse
it, any more than a project on SourceForge requires endorsement by Linus
if it runs on Linux ;-)
(Or, if you prefer, don't keep whingeing about what the MPI Forum chose
to do, but get on and fix it for yourself).
James Cownie <jcownie at etnus.com>
Etnus, LLC. +44 117 9071438
More information about the Beowulf