On Jan 15, 2008, at 3:57 AM, Lenny Verkhovsky wrote:
Ah, you meant the listener socket -- not 2 sockets to each peer. Ok,
fair enough. Opening up one more listener socket in each process is
no big deal (IMO).
I thought in a large cluster it can be a problem.
If you're adding one more listener socket in each process, that's no
big deal. That's an O(1) increase -- not an O(N) (or O(N^2)) increase.
Perhaps I don't understand exactly what you are proposing. I was
under the impression that you were going after a common case: mpirun
and the MPI jobs are running on back-end compute nodes where all of
them support SDP (although the other case of mpirun running on the
head node without SDP and all the MPI processes are running on back-
end nodes with SDP is also not-uncommon...). Are you thinking of
something else, or are you looking for more flexibility?
I am just looking for more flexibility for the end user.
Can you cite some specific use-cases that you want to be able to
support? I think we're talking around in circles because we're
talking about different scenarios.
--
Jeff Squyres
Cisco Systems
