On 06/01/2012 05:06 PM, Edmund Sumbar wrote:
Thanks for the tips Gus. I'll definitely try some of these, particularly the nodes:ppn syntax, and report back.
You can check for torque support with mpicc --showme It should show among other things -ltorque [if it has torque support] and -lrdmacm -libverbs [if it has OpenIB/Infinband support]. If Torque is not installed in a standard location [such as /usr or /usr/local], which is often the case, you may need to point configure to the Torque library with: --with-tm=/path/to/torque Likewise for Infinband/OpenIB if you have it: --with-openib=/path/to/openib [I am citing these options from memory. Do a './configure -help' to check the right syntax, please.] Making a log file of your configure run may be helpful, to diagnose problems. Finally, if I remember right, there was some problem reported in the list regarding Intel compilers 12.1. [I built 1.4.5 with Intel 11 and it works fine.] However, that problem may have been superseded in the latest OpenMPI 1.6.0. [The release notes will tell, or perhaps Jeff.] I hope this helps, Gus Correa
Right now, I'm upgrading the Intel Compilers and rebuilding Open MPI. On Fri, Jun 1, 2012 at 2:39 PM, Gus Correa <g...@ldeo.columbia.edu <mailto:g...@ldeo.columbia.edu>> wrote: The [Torque/PBS] syntax '-l procs=48' is somewhat troublesome, and may not be understood by the scheduler [It doesn't work correctly with Maui, which is what we have here. I read people saying it works with pbs_sched and with Moab, but that's hearsay.] This issue comes back very often in the Torque mailing list. Have you tried instead this alternate syntax? '-l nodes=2:ppn=24' [I am assuming here that your nodes have 24 cores, i.e. 24 'ppn', each] Then in the script: mpiexec -np 48 ./your_program Also, in your PBS script you could print the contents of PBS_NODEFILE. cat $PBS_NODEFILE A simple troubleshooting test is to launch 'hostname' with mpirun mpirun -np 48 hostname Finally, are you sure that the OpenMPI you are using was compiled with Torque support? If not, I wonder if clauses like '-bynode' would work at all. Jeff may correct me if I am wrong, but if your OpenMPI lacks Torque support, you may need to pass to mpirun the $PBS_NODEFILE as your hostfile. -- Edmund Sumbar University of Alberta +1 780 492 9360 _______________________________________________ users mailing list us...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/users
