The following provides a basic introduction to getting started with Slurm. Users will learn about the Slurm resource manager/job scheduler, how to specify the resources they need, and how to submit, monitor, and cancel their jobs on the Compute node.
Get started with Slurm
In order to request cluster resources through Slurm, you’ll need to create an Sbatch script file, also called a Slurm script. Your Sbatch scripts should generally follow the format below:
#!/bin/bash
# Declaring Slurm Configuration Options
# Loading Software/Libraries
# Running Code
The first line indicates the program to run the Slurm script file with. In this case, it’s bash. You’re recommended to use bash. If you use tcsh, you can define the first line as: #!/bin/tcsh