Skip to content
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.

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

Multi-node Execution Support in Sarek on HPC Systems #1768

Closed
anya-shm opened this issue Jan 8, 2025 · 2 comments
Closed

Multi-node Execution Support in Sarek on HPC Systems #1768

anya-shm opened this issue Jan 8, 2025 · 2 comments

Comments

@anya-shm
Copy link

anya-shm commented Jan 8, 2025

Hello Sarek team,

I am currently running the nf-core Sarek pipeline (v3.5.0) on an HPC system and encountered an issue when attempting to utilize multiple nodes. I launched the pipeline with the following PBS settings:

#PBS -l nodes=3:ppn=8

This should provide a total of 24 CPUs (3 nodes × 8 CPUs per node). However, I received the following error during execution:

Process requirement exceeds available CPUs -- req: 24; avail: 8

It appears that the pipeline is only recognizing the CPUs on a single node rather than across all allocated nodes. I would like to ask:

  • Does Sarek support multi-node execution when run on HPC systems?
  • If supported, are there specific configurations or additional parameters required to enable multi-node execution?

Any guidance or documentation on how to resolve this issue would be greatly appreciated.

Thank you for your help!

Best regards,
Anna.

@FriederikeHanssen
Copy link
Contributor

Do you mean with multi-node execution that a single task is run on two nodes? If so, then I don't believe nextflow and by extension sarek supports it.

The above error indicates that you are requesting 24CPUs for a single task but there are only nodes with 8 available. To avoid this error you can use [resourceLimits](https://www.nextflow.io/docs/latest/reference/process.html#resourcelimits) and set them to the maximal allowed resources an individual task can consume,

@anya-shm
Copy link
Author

anya-shm commented Jan 8, 2025

Thank you for the quick reply and clarification!

Yes, I meant multi-node for a single process. I will try with resourceLimits, as you suggested.

Do you mean with multi-node execution that a single task is run on two nodes? If so, then I don't believe nextflow and by extension sarek supports it.

The above error indicates that you are requesting 24CPUs for a single task but there are only nodes with 8 available. To avoid this error you can use [resourceLimits](https://www.nextflow.io/docs/latest/reference/process.html#resourcelimits) and set them to the maximal allowed resources an individual task can consume,

@anya-shm anya-shm closed this as completed Jan 8, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants