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

Create an example of how to use nodeSelector and nodeAffinity to request specific GPU types #598

Closed
supertetelman opened this issue Aug 6, 2020 · 1 comment
Labels

Comments

@supertetelman
Copy link
Collaborator

@supertetelman supertetelman commented Aug 6, 2020

Since we are planning to include the GPU Feature Discovery in a future release of DeepOps it might be useful to include a few examples of how to use GPU Node Labels to request specific GPU types, CUDA versions, etc.

I haven't found any solid documentation or examples on how to do this, so if they exist, perhaps simply linking to them would be good enough in the README, if not we should create:

  • How to request a specific GPU family
  • How to restrict a specific GPU family
  • How to request a minimal driver version
  • How to request MIG in "single" mode

This might be a better feature request for the actuo GPU Feature Discovery repo (https://github.com/NVIDIA/gpu-feature-discovery).

@supertetelman
Copy link
Collaborator Author

@supertetelman supertetelman commented Nov 10, 2020

Some examples of NodeSelectors can now be found here: https://github.com/NVIDIA/deepops/blob/master/docs/k8s-cluster/nvidia-mig.md.

We could use a few more examples of NodeAffinity with minimal CUDA versions, etc.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Linked pull requests

Successfully merging a pull request may close this issue.

None yet
1 participant
You can’t perform that action at this time.