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

Sharding support #322

Closed
wants to merge 34 commits into from
Closed

Sharding support #322

wants to merge 34 commits into from

Conversation

etiennedub
Copy link
Contributor

depends ComputeCanada/magic_castle#289
close #299

Puppet configuration to support sharding

Activate autodetect=nvml only on the compute nodes and generate
a static gres.conf file for the controller based on the content
of terraform_data.yaml
This is the only version compiled with NVML for now
slurmd should only be started if all running conditions are fulfilled
and Puppet is the only actor that knows about all of the conditions.
We therefore make Puppet the only actor that can start slurmd.
The reboot program is only called once the node is idle, making it
a great choice to reboot nodes when jobs are running.
The mig config is managed by Puppet only.
Use the class name tag to notify slurmd on exec
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.

2 participants