Job directives

Summary

Slurm equivalent job specification ex. --node=m01

Body

Option MOAB Torque command (mksub) Slurm command (sbatch)
Job Name -N <name> --job-name=<name>
Account -a <account> --account=<account>
Queue -q <queue> --partition=<name>
Node Count nodes=<number of nodes> --nodes=<count>
Process count per node ppn=<number of cores> --ntasks-per-node=<count>
Core Count (per process)   --cpus-per-task=<cores>
Wall time limit walltime=<hh:mm:ss> --time=<hh:mm:ss>
GPU request -gpus=<count> --gres=gpu:<count>
Memory limit   --mem=<limit> (in MB)
Standard output file -o=<file path> --output=<file path>
Standard error file -e=<file path> --error=<file path>
Combine stdout/stderr to stdout -j oe --output=<combined out and err file path>
Job array -t <number of array indices> --array=<array indices>
Copy environment   --export=ALL (default --export=NONE, no export)
Job dependency   --dependency=after:jobID[:jobID...]
--dependency=afterok:jobID[:jobID...]
--dependency=afternotok:jobID[:jobID...]
--dependency=afterany:jobID[:jobID...]
Job status notification -m bea --mail-type=BEGIN,END,FAIL
Email address -M <email address> --mail-user=<email address>
Append to job file   --open-mode=append.
Specify a compute node hostlist=<name of compute node> --nodelist=<name of compute node>
Email before job termination qnotify; #SBATCH --signal=B:USR1@70;

Details

Details

Article ID: 132871
Created
Tue 6/1/21 10:29 AM
Modified
Tue 3/8/22 10:22 AM