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.

Sign up for GitHub

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

Jump to bottom

Support kubeflow operator #297

Closed
1 of 2 tasks
Tracked by #636
, #974
...
xiaoxubeii opened this issue Jul 14, 2022 · 12 comments · Fixed by #1183
Closed
1 of 2 tasks
Tracked by #636
, #974
...

Support kubeflow operator #297

xiaoxubeii opened this issue Jul 14, 2022 · 12 comments · Fixed by #1183
Assignees
@tenzen-y
Labels
kind/feature Categorizes issue or PR as related to a new feature. lifecycle/frozen Indicates that an issue or PR should not be auto-closed due to staleness.

Comments

@xiaoxubeii
Copy link

xiaoxubeii commented Jul 14, 2022 •
edited
Loading

What would you like to be added:
Support kubeflow training operator.

Why is this needed:
It is to track the status of kueue to support kubeflow training operator.
@xiaoxubeii xiaoxubeii added the kind/feature Categorizes issue or PR as related to a new feature. label Jul 14, 2022
@alculquicondor
Copy link
Contributor

Note that MPIJob latest version is not currently part of the training-operator kubeflow/training-operator#1479

@k8s-triage-robot
Copy link

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

  • After 90d of inactivity, lifecycle/stale is applied
  • After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
  • After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

  • Mark this issue or PR as fresh with /remove-lifecycle stale
  • Mark this issue or PR as rotten with /lifecycle rotten
  • Close this issue or PR with /close
  • Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

@k8s-ci-robot k8s-ci-robot added the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Oct 12, 2022
@alculquicondor
Copy link
Contributor

/lifecycle frozen

@k8s-ci-robot k8s-ci-robot added lifecycle/frozen Indicates that an issue or PR should not be auto-closed due to staleness. and removed lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. labels Oct 12, 2022
@tenzen-y tenzen-y mentioned this issue Mar 20, 2023
Closed
@alculquicondor
Copy link
Contributor

This is currently blocked on kubeflow/common#196

@alculquicondor alculquicondor mentioned this issue Mar 29, 2023
Closed
@tenzen-y
Copy link
Member

tenzen-y commented Jul 5, 2023

/assign

This was referenced Jul 11, 2023
Closed
Merged
@tenzen-y
Copy link
Member

tenzen-y commented Jul 19, 2023 •
edited
Loading

As a first step, I opened a PR to add the PyTorchJob support, and then I will add the following framework support:

  • TFJob
  • MXJob
  • XGboostJob
  • PaddleJob

Also, I'm on the fence if we should support MPIJob v1 hosted only on kubeflow/training-operator (currently, MPIJob v2 hosted only on kubeflow/mpi-operator)

Regarding MPIJob v1 wdyt? @alculquicondor @mimowo @kerthcet @trasc

@alculquicondor
Copy link
Contributor

I'm ok leaving it out if it's not trivial to support 2 API versions. I think the CRD objects themselves are not compatible.

@tenzen-y
Copy link
Member

I think the CRD objects themselves are not compatible.

Right.

I'm ok leaving it out if it's not trivial to support 2 API versions.

We can not support v1 and v2 API by a single controller: https://github.com/kubernetes-sigs/kueue/tree/a103723023aa6c5a63cc8c1248fd38d8640d7003/pkg/controller/jobs/mpijob.

However, once we implement a separate controller for v1 like https://github.com/kubernetes-sigs/kueue/blob/3589969054023cb8b584a4639f4b9dec8c371a67/pkg/controller/jobs/kubeflow/jobs/pytorchjob/pytorchjob_controller.go, we can support v1.

@tenzen-y
Copy link
Member

Anyway, I think MPIJob v1 is a lower priority since we already support MPIJob v2.

@kerthcet
Copy link
Contributor

kerthcet commented Jul 20, 2023 •
edited
Loading

+1 to defer the work unless we receive strong demands.

@tenzen-y
Copy link
Member

+1 to defer the work unless we receive strong demands.

I agree.

@tenzen-y
Copy link
Member

tenzen-y commented Aug 16, 2023 •
edited
Loading

Tasks:

@tenzen-y tenzen-y mentioned this issue Aug 17, 2023
Merged
@tenzen-y tenzen-y mentioned this issue Sep 12, 2023
Merged
@tenzen-y tenzen-y mentioned this issue Sep 21, 2023
Merged
@tenzen-y tenzen-y mentioned this issue Oct 9, 2023
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@tenzen-y tenzen-y

Labels
kind/feature Categorizes issue or PR as related to a new feature. lifecycle/frozen Indicates that an issue or PR should not be auto-closed due to staleness.
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Support kubeflow.org/mxjob tenzen-y/kueue
6 participants
@xiaoxubeii @alculquicondor @kerthcet @k8s-ci-robot @k8s-triage-robot @tenzen-y