Skip to content

Conversation

@elena-buscaroli
Copy link
Contributor

@elena-buscaroli elena-buscaroli commented Jan 21, 2026

PR checklist

Closes #9724

The PR changes avoids fitting the tool in parallel.

  • This comment contains a description of changes (with reason).
  • If you've fixed a bug or added code that should be tested, add tests!
  • If you've added a new tool - have you followed the module conventions in the contribution docs
  • If necessary, include test data in your PR.
  • Remove all TODO statements.
  • Broadcast software version numbers to topic: versions - See version_topics
  • Follow the naming conventions.
  • Follow the parameters requirements.
  • Follow the input/output options guidelines.
  • Add a resource label
  • Use BioConda and BioContainers if possible to fulfil software requirements.
  • Ensure that the test works with either Docker / Singularity. Conda CI tests can be quite flaky:
    • For modules:
      • nf-core modules test <MODULE> --profile docker
      • nf-core modules test <MODULE> --profile singularity
      • nf-core modules test <MODULE> --profile conda
    • For subworkflows:
      • nf-core subworkflows test <SUBWORKFLOW> --profile docker
      • nf-core subworkflows test <SUBWORKFLOW> --profile singularity
      • nf-core subworkflows test <SUBWORKFLOW> --profile conda

@elena-buscaroli
Copy link
Contributor Author

elena-buscaroli commented Jan 21, 2026

Hi @itrujnara! Since you approved another PR I published earlier (#9727) which is really similar to this one, I suggested you as the reviewer. If you are available, thanks!

Copy link
Contributor

@SPPearce SPPearce left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why do you NOT want to fit the model in parallel?

@elena-buscaroli
Copy link
Contributor Author

Hi @SPPearce! Parallelisation is implemented via the external R package easypar, which internally relies on foreach/doParallel to spawn and distribute independent tasks across multiple worker processes, with the number of workers inferred from the available cores. This is not explicitly designed to integrate with HPC schedulers and does not account for cluster-specific resource allocation mechanisms. As a result, we have observed inconsistent behaviour and resource-related issues on some HPC systems, resulting in unexpected crashes of the module.

Since the module processes only one patient at a time, the runtime is relatively short (on the order of minutes), we are assigning the process_single label, which requests a single CPU and we believe it is safer to disable internal parallelisation in order to avoid potential cluster-specific issues.

Let me know what you think about it, thanks!

@elena-buscaroli elena-buscaroli added this pull request to the merge queue Jan 22, 2026
Merged via the queue into nf-core:master with commit f6031b1 Jan 22, 2026
28 checks passed
@elena-buscaroli elena-buscaroli deleted the viber branch January 22, 2026 10:02
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

Successfully merging this pull request may close these issues.

update module: viber

2 participants