SLURM Overview
Rockfish uses SLURM (Simple Linux Universal Resource Manager) to manage resource scheduling, job submission, and execution across the cluster.
SLURM is a widely adopted, open-source workload manager developed by SchedMD. It supports large-scale high-performance computing environments and is used by national labs, supercomputing centers, and universities worldwide.
Overview
SLURM allows users to:
Submit batch jobs using sbatch
Run interactive sessions using interact
Monitor job status with squeue, sacct, or scontrol
Manage job arrays, resource requests, and job dependencies
Allocate compute, memory, and GPU resources efficiently
All compute-intensive jobs must be submitted through SLURM. Running jobs directly on the login nodes is strictly prohibited. These nodes are shared among all users and are reserved for lightweight tasks like editing scripts, submitting jobs, and checking output files.
Rockfish places limits on the number of jobs queued and running on a per group (allocation) and partition basis. Please note that submitting a large number of jobs (especially very short ones) can impact the overall scheduler response for all users. If you are anticipating submitting a lot of jobs, please contact help@rockfish.jhu.edu before submission. We can work to check if there are bundling options (job arrays or gnu parallel) that make your workflow more efficient and reduce the impact on the scheduler.
Interactive Sessions
If your workflow requires interacting with a job as it runs (e.g., debugging, live monitoring, or launching GUI applications like RStudio or Jupyter), you must request an interactive session using the interact command:
interact -p shared -n 4 -t 02:00:00
This submits a job to the scheduler that, once scheduled, grants you interactive terminal access to a compute node.
Learn More
To learn more about SLURM commands and features, refer to the official documentation: