Submitting jobs

All jobs run on the HPC systems should be submitted via the queueing system, Slurm.

Typically, users submit jobs to Slurm by preparing a non-interactive job script that specifies the required resources and lists the commands to execute on those resources.

These pages provide an overview of the Slurm components available on the cluster, and instructions for submitting jobs with specified resources.

Users who are new to batch queue systems or Slurm should start with the Slurm overview page, which covers the terminology used throughout this site before reading other pages.

See the following pages for more information:

• Slurm overview
• The job script
• Memory
• Runtime
• Partitions
• Single node jobs
• Multi-node jobs
• High memory jobs
• Interactive jobs
• Constraints
• Job output
• Common options