Closed adbeggs closed 2 months ago
Slurm is supported through Nextflow's integrations with slurm. The precise config file used is intimately reliant on the end-user's HPC systems. Its not something we can support to include configuration files for all our users (and neither do we condone the idea of people submitting private cluster config profiles to a public respository).
Hi Chris
That’s kind of what Nextflow already support here:
https://github.com/nf-core/configs
Looking at your Nextflow config, I would have to rewrite your pipeline to explicitly support Slurm as it is not mentioned in the Nextflow.config and so doesn’t support it out of the box (unlike nf-core)
Happy to do this myself
A
From: Chris Wright @.> Sent: Tuesday, April 23, 2024 1:27 PM To: epi2me-labs/wf-human-variation @.> Cc: Andrew Beggs (Cancer and Genomic Sciences) @.>; Author @.> Subject: Re: [epi2me-labs/wf-human-variation] Explicit support for SLURM and other schedulers (Issue #176)
CAUTION: This email originated from outside the organisation. Do not click links or open attachments unless you recognise the sender and know the content is safe.
Slurm is supported through Nextflow's integrations with slurm. The precise config file used is intimately reliant on the end-user's HPC systems. Its not something we can support to include configuration files for all our users (and neither do we condone the idea of people submitting private cluster config profiles to a public respository).
— Reply to this email directly, view it on GitHubhttps://github.com/epi2me-labs/wf-human-variation/issues/176#issuecomment-2072167433, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AC7KTDFEQTPX5B4RO4E6XNDY6ZHPZAVCNFSM6AAAAABGUXF4P2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZSGE3DONBTGM. You are receiving this because you authored the thread.Message ID: @.***>
We're very aware of https://github.com/nf-core/configs, this is what we find it hard to condone. Many IT professionals would consider information in that repository a security risl for their infrastructure.
You don't need to rewrite the pipeline. You need only to provide an additional config file containing the configuration required.
Is your feature related to a problem?
Although you can submit to a slurm cluster, the subjobs are not submitted using SLURM meaning that it is less efficient
Describe the solution you'd like
Explicit support of job schedulers in tasks, along the line of NF-Core
Describe alternatives you've considered
Doing it myself, but I don't have the time!
Additional context
No response