Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

Commands Cheatsheet

View all jobs in the queue

squeue

View all partitions

sinfo -a

View partitions user can access (SLURM accounting)

sacctmgr show associations where user=$USER

You may also want see a list of all of the partitions if the none show up in the previous command.

sinfo -a

(RCC partitions: https://docs.rcc.uchicago.edu/partitions/)

Launch an interactive CPU job (example)

RCC

srun --account=<PI_ACCOUNT> --partition=caslake --nodes=1 --ntasks=1 --cpus-per-task=2 --mem=4G --time=01:00:00 --pty bash -i

DSI

srun --partition general --nodes=1 --ntasks=1 --cpus-per-task=2 --mem=4GB --time=01:00:00 --pty bash -i

Launch an interactive GPU job (example)

srun -p general --gres=gpu:a100:1 --mem=32G -t 01:30:00 --pty /bin/bash

Note: You may have to locate how to reference available GPUs: View node and GPU information

View your jobs

squeue -u $USER

Cancel a job

scancel <job_id>

Show job details

scontrol show job $SLURM_JOB_ID

View node and GPU information

sinfo -o "%50N %10c %20m %30G" NODELIST CPUS MEMORY GRES

Attach to a running job

srun --jobid=<JOBID> --pty /bin/bash

Check GPU status (NVIDIA GPUs)

nvidia-smi
User Guide
Quick Reference
User Guide
Data Transfer