[MPICH] mpiexec -soft with Sun Grid Engine

[Rajeev Thakur]

I don't know about Sun Grid Engine, but on a regular machine, it seems to
work. Here is what I get:

mpich2-1.0.3/examples% mpiexec -n 5 cpi
Process 0 of 5 is on shakey.mcs.anl.gov
Process 1 of 5 is on shakey.mcs.anl.gov
Process 2 of 5 is on shakey.mcs.anl.gov
Process 3 of 5 is on shakey.mcs.anl.gov
Process 4 of 5 is on shakey.mcs.anl.gov
pi is approximately 3.1415926544231230, Error is 0.0000000008333298
wall clock time = 0.014919

mpich2-1.0.3/examples% mpiexec -n 5 -soft 2 cpi
Process 0 of 2 is on shakey.mcs.anl.gov
Process 1 of 2 is on shakey.mcs.anl.gov
pi is approximately 3.1415926544231318, Error is 0.0000000008333387
wall clock time = 0.002462

mpich2-1.0.3/examples% mpiexec -n 5 -soft 0:4 cpi
Process 0 of 4 is on shakey.mcs.anl.gov
Process 1 of 4 is on shakey.mcs.anl.gov
Process 2 of 4 is on shakey.mcs.anl.gov
Process 3 of 4 is on shakey.mcs.anl.gov
pi is approximately 3.1415926544231239, Error is 0.0000000008333307
wall clock time = 0.010566

mpich2-1.0.3/examples% mpiexec -n 5 -soft 0:5 cpi
Process 0 of 5 is on shakey.mcs.anl.gov
Process 1 of 5 is on shakey.mcs.anl.gov
Process 2 of 5 is on shakey.mcs.anl.gov
Process 3 of 5 is on shakey.mcs.anl.gov
Process 4 of 5 is on shakey.mcs.anl.gov
pi is approximately 3.1415926544231230, Error is 0.0000000008333298
wall clock time = 0.015353
  

> -----Original Message-----
> From: owner-mpich-discuss at mcs.anl.gov 
> [mailto:owner-mpich-discuss at mcs.anl.gov] On Behalf Of Yusong Wang
> Sent: Friday, July 21, 2006 11:02 AM
> To: mpich-discuss at mcs.anl.gov
> Subject: [MPICH] mpiexec -soft with Sun Grid Engine
> 
> Hi,
> 
> I am trying to take advantage of the "-soft" option on a busy cluster.
> In the Appendix D (P362) of "Using MPI-2":
> mpiexec -n 10 -soft 2:10 program
> will run the program on various number of processes, depending on what
> is available.
> 
> But this option doesn't work at all on our system, even 
> though there are
> enough nodes for the lower bound in the soft option. The version of
> MPICH2 is: mpich2_mpd_sock v 1.78 2005/09/23
> 
> I read the python source code for mpiexec and tried:
> mpiexec -n 10 -soft 2 program
> It does not work either.
> 
> I wonder what is available for this option in the current distribution
> of MPICH2. Also, should mpiexec talk to Sun Grid Engine 
> periodically to
> see if there is enough resource to run the job with the lower bound
> number of CPUs after submission? Or just check the 
> availability of CPUs
> once at submission.
> 
> Thanks,
> 
> Yusong
> 
>