Skip to content

add QOS-based GPU allocation info for job submissions #91

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
dlevenstein merged 1 commit into from
Mar 16, 2026
+3 −0
Merged

add QOS-based GPU allocation info for job submissions #91

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions docs/Resources/hpc.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,6 +49,9 @@ To get started visit this page for [some info](https://docs.ycrc.yale.edu/cluste
5. `mpi`: for highly-parallelized code
6. `pi_NAME`: PI and lab specific nodes available for purchace from YCPC

### QOS-Based GPU Allocation
By default, jobs are submitted with standard priority. This means that when cluster resources are scarce, these jobs can be preempted, or terminated, in favor of high-priority jobs. This isn't generally a problem for shorter jobs, but for long jobs that are resource-intensive, it may make sense to submit a high-priority job using a SLURM mechanism called QOS (Quality of Service). Our lab has a certain number of guaranteed resources which you can gain access by using the `--qos=qos_levenstein` along with `salloc` or `sbatch`. See the [Misha Guide](https://ood-misha.ycrc.yale.edu/public/misha/04-qos.html) for more info. It's also recommended to use checkpoints in case a job does get terminated.

### Cheat Sheet

* Interactive jobs:
Expand Down