[MPICH] mpirun vs. mpiexec

[Steve Young]

I just tried some tests with that version of mpiexec and it is now
working much better. Thanks a lot for the help. 


On Tue, 2007-06-05 at 14:18 -0500, Anthony Chan wrote:
> 
> On Tue, 5 Jun 2007, Matthew Chambers wrote:
> 
> >
> > > > I was contemplating switching over to the setup where each user starts
> > > > their own ring. I have an older cluster that I am planning on testing
> > > > this setup with. I am curious to know if there are
> > > > advantages/disadvantages over the two setups. Would I notice anything
> > > > different in terms of performance?
> > >
> > > I don't think you will see any difference in MPI performance whether mpds
> > > are started by root or users...
> >
> > If one is using a lot nodes, doesn't setting up the MPI ring incur quite a
> > bit of overhead with the ssh calls?
> >
> 
> I am refering to the MPI communication performance, not the startup cost
> of setting up the mpd ring.  Besides setting 38 nodes of mpds with ssh
> is fast.  Of course, if you are talking about thousand of nodes, Pete
> Wyckoff's mpiexec is probably the best option as pbs_mom is already
> running on each compute node....
> 
> A.Chan