Event Details:
Wednesday, April 16, 2025
2:00pm - 3:00pm PDT
Learn how to help reduce your wait in the job queue and more efficiently utilize CPUs, GPUs and memory. This session will cover the basics of estimating job RAM and CPU needs with SLURM. We will go over several methods and tools in this presentation:
- SLURM's sacct command
- htop
- nvtop
- ruse
- ssh'ing to monitor a running job
The presentation's slide deck is here