User Tools

Site Tools


strutture:cnaf:clusterhpc:using_the_cnaf_hpc_cluster

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
strutture:cnaf:clusterhpc:using_the_cnaf_hpc_cluster [2017/07/04 11:06]
dcesini@infn.it [Requesting Access]
strutture:cnaf:clusterhpc:using_the_cnaf_hpc_cluster [2021/03/05 14:54]
dcesini@infn.it [Alternative MPI multinode submission]
Line 126: Line 126:
 </code> </code>
  
-==== Submitting MPI Jobs ====+==== Submitting MPI Jobs via mpirun.lsf (obsolete)====
  
 Currently only **OpenMPI** jobs have been tested on the HPC cluster. \\ Currently only **OpenMPI** jobs have been tested on the HPC cluster. \\
Line 187: Line 187:
 (Thanks to S.Sinigardi for sharing it)\\ (Thanks to S.Sinigardi for sharing it)\\
  
 +
 +==== Alternative MPI multinode submission ====
 +It is possible to avoid the usage of mpirun.lsf and dinamically set the mpirun machine file in the following way:
 +
 +1) Create automatically the machine file to be using in the mpirun:
 +
 +    echo $LSB_HOSTS | awk '{split($0,array," ")} END {for (i in array) printf ("%s\n",array[i])}' | awk '{count[$0]++} END {for (word in count) print word,"slots=" count[word]}' > /home/HPC/username/mymachine.txt
 +
 +2) Use this command to launch mpirun:
 +
 +    mpirun --machinefile /home/HPC/username/machinefile.txt -x PSM_SHAREDCONTEXTS_MAX=8 -np $LSB_DJOB_NUMPROC /home/HPC/username/executablename
 +
 +A possible bsub submission is:  
 +
 +    bsub -q hpc_inf_SL7  -n 16 -R "span[ptile=8]" -o testmpimy.out -e testmpimy.err /home/HPC/username/run_this_example.sh
 +
 +where in the run_this_example.sh script you launch the previous commands:
 +
 +----run_this_example.sh----
 +
 +    #!/bin/bash
 +
 +    echo $LSB_HOSTS | awk '{split($0,array," ")} END {for (i in array) printf ("%s\n",array[i])}' | awk '{count[$0]++} END {for (word in count) print word,"slots=" count[word]}' > /home/HPC/username/mymachine.txt
 +
 +    mpirun --machinefile /home/HPC/username/machinefile.txt -x PSM_SHAREDCONTEXTS_MAX=8 -np $LSB_DJOB_NUMPROC /home/HPC/username/executablename
 ==== Submitting GPU Jobs ==== ==== Submitting GPU Jobs ====
  
strutture/cnaf/clusterhpc/using_the_cnaf_hpc_cluster.txt ยท Last modified: 2021/03/05 14:54 by dcesini@infn.it