Scheduler Cheatsheet for AWS ParallelCluster

schedulers.md

General

	sge	slurm	torque
Submit Interactive Job	qlogin	srun	qsub -I
Submit Batch Job	qsub	sbatch	qsub
Number of Slots	-pe mpi [n]	-n [n]	-l ppn=[n]
Number of Nodes	-pe mpi [slots * n]	-N [n]	-l nodes=[n]
Cancel Job	qdel	scancel	qdel
See Queue	qstat	squeue	qstat
See Nodes	qhost	sinfo -N	pbsnodes
Script Directive	#$	#SBATCH	#PBS
Job Name	-N [name]	--job-name [name]	-N [name]

Slurm

Submit jobs with:

$ sbatch
$ srun

See the queue with:

$ squeue

Cancel Jobs with:

$ scancel [jobid]

See the nodes details:

$ sinfo -Nl
Fri May  8 16:49:47 2020
NODELIST        NODES PARTITION       STATE CPUS    S:C:T MEMORY TMP_DISK WEIGHT AVAIL_FE REASON              
ip-10-0-10-70       1  compute*        idle   72   72:1:1      1        0      1   (null) none                
ip-10-0-10-133      1  compute*        idle   72   72:1:1      1        0      1   (null) none    

Example job submission script:

#!/bin/bash
#SBATCH --job-name=test
#SBATCH --ntasks=8
#SBATCH --output=%x_%j.out
module load openmpi
mpirun job

Run Job on All Nodes

SGE

NODES=30
CORES=36
qsub -t 1-$NODES:1 -pe mpi $NODES*$CORES job.sh

SLURM

NODES=30
srun --ntasks-per-node 1 --ntasks $NODES job.sh

TORQUE

qsub -l nodes=$NODES:ppn=1 job.sh