Basic Slurm Commands

Salloc

Salloc runs an interactive job on the cluster. You can request a shell to run commands or submit a script or app interactively.

Example usage:

salloc --nodes=1 --ntasks=1 --cpus-per-task=8 --time=1:00:00 --partition=a100-xl --gres=gpu:1 --mem=16GB --job-name=interactive-test

The command line arguments describe the job configuration, and are listed in detail on the slurm website.

See full salloc docs on the slurm website

Sbatch

Sbatch submits a script to be run as a batch job on the cluster. This allocates a job, similar to salloc, but you do not get a shell into the job and cannot (easily) control the job once started.

Before running sbatch, you must define a script. E.g. nano myjob and paste the following text:

#!/bin/bash
#SBATCH --nodes=1
#SBATCH --tasks=1
#SBATCH --cores-per-task=8
#SBATCH --time=04:00:00
#SBATCH --partition=a100-xl
#SBATCH --mem=16GB
#SBATCH --job-name=test-sbatch
#SBATCH --output=myjob_%j.out
#SBATCH --error=myjob_%j.err
#SBATCH --gres=gpu:1

# do stuff here ...

The #SBATCH lines describe paramters used to allocate and configure the job. E.g. #SBATCH --nodes=1 tells the scheduler to allocate one node for the job. These are mostly identical to the command line options for salloc. output and error list files where STDOUT and STDERR will be written, respectively. The symbol %j in the output and error file names will be replaced by the job number.

To submit the myjob script you wrote, run the following command:

sbatch myjob

See full sbatch docs on the slurm website

Squeue

Squeue is used to see the slurm queue. This include jobs which are currently running, pending, configuring, etc. It will not show jobs which have completed or failed.

See full squeue docs on the slurm website

Example usage:

squeue

You can show jobs for a specific user with

squeue --user=$USER

#or, to show jobs for your own account

squeue --me

Sinfo

Sinfo shows partition and node information.

You can list partitions using:

sinfo

See full sinfo docs on the slurm website

Sacct

Sacct shows accounting and historical information for slurm jobs. Unlike squeue which only shows active jobs, sacct allows you to get information about jobs which have completed or failed.

E.g. get information about a specific job

sacct -j 1234

where 1234 is replaced by the job ID

Useful snippets

List detailed information about a job:

scontrol show jobid -dd $YOUR_JOBID_HERE

Cancel all running/queued jobs that you have submitted:

scancel -u `whoami`

See full sacct docs on the slurm website