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

[Feature] Generate rayStartParam resource flag from container resource(e.g. cpu,memory) #152

Closed
2 tasks done
chenk008 opened this issue Feb 17, 2022 · 4 comments
Closed
2 tasks done
Labels
enhancement New feature or request

Comments

@chenk008
Copy link
Contributor

Search before asking

  • I had searched in the issues and found no similar feature requirement.

Description

Using container.resource.cpu.limit as num-cpus default value in rayStartParams, it will help to reduce configuration redundancy. When we want to change the container resource config, we don't need to update rayStartParams.

It's the same with memory,gpu.

Use case

No response

Related issues

No response

Are you willing to submit a PR?

  • Yes I am willing to submit a PR!
@chenk008 chenk008 added the enhancement New feature or request label Feb 17, 2022
@Jeffwan
Copy link
Collaborator

Jeffwan commented Feb 17, 2022

Agree. container.resource.cpu.limit should be the source the truth. Let's do this

@Jeffwan Jeffwan mentioned this issue Mar 4, 2022
3 tasks
@ebr
Copy link
Contributor

ebr commented Mar 9, 2022

Perhaps this should be set from requests instead of limits? It's sometimes beneficial to not set the CPU limit, as it can introduce unwanted throttling.

@Jeffwan
Copy link
Collaborator

Jeffwan commented Mar 10, 2022

I think ray is not aware of the (limit-request) resources which leads wasting? @chenk008 do you know exact behavior on ray side?

@Jeffwan
Copy link
Collaborator

Jeffwan commented May 1, 2022

It has been addressed in issue #170 and we can close the issue

@Jeffwan Jeffwan closed this as completed May 1, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

3 participants