Status of kube-batch

[Huang-Wei]

Per requirement from the Kubernetes steering committee, we'd like to make the latest status of each sub-project clearly documented to the public.

@k82cn I suppose volcano is the successor of kube-batch, and hence kube-batch is no longer actively maintained. So should we either document the status in README, or as you requested earlier, make kube-batch archived?

[k82cn]

I'm going to update README, and then let's archived this repo :)

[Huang-Wei]

I'm going to update README, and then let's archived this repo :)

Thanks Klaus!

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough active 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 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 rotten

[k8s-triage-robot]

The Kubernetes project currently lacks enough active 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:

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

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

/close

[k8s-ci-robot]

@k8s-triage-robot: Closing this issue.

In response to this:

The Kubernetes project currently lacks enough active 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:

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

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

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[Huang-Wei]

/reopen

@k82cn any plan on maintaining the health/activeness of this project?

[k8s-ci-robot]

@Huang-Wei: Reopened this issue.

In response to this:

/reopen

@k82cn any plan on maintaining the health/activeness of this project?

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[Huang-Wei]

/remove-lifecycle rotten

[k82cn]

@k82cn any plan on maintaining the health/activeness of this project?

@Huang-Wei , please feel free to archive this one, I put more time on other project, e.g. https://github.com/xflops/flame .

[kerthcet]

Hi I hope to put some efforts to this project, can you defer the archive. Or I can help to maintain this project, I'm currently working on the batch workloads and hope to make it works smoothly with the kubernetes, especially with the kube-scheduler.

[kerthcet]

By the way, I'm a reviewer of kubernetes sig scheduling, and a developer of volcano, also I work on the batch wg to contribute ideas and solutions to the upstream, e.g. kueue(a kubernetes native job queueing system), I hope kubernetes can support the batch scenarios smoothly.

[Huang-Wei]

@kerthcet i may prefer to star a new subproject if needed. This repo's codebase may beed big revamp to catch up with latest k8s version, which IMO may not worth the efforts.

[kerthcet]

Thanks @Huang-Wei I'll reconsider your suggestion about sponsoring a new project.

[k82cn]

. This repo's codebase may beed big revamp to catch up with latest k8s version, which IMO may not worth the efforts.

The codebase is not bound to k8s version, it'll not take much effort to bump up the version.

@kerthcet , why not contribute to Volcano? And would you share your use case for kube-batch?

[kerthcet]

Volcano is an excellent project, our company is exploring it right now, so I'm also contributing to it. But it has its own ecosystem, it has its own job api, job queueing system, but we're now supporting it upstream, although not as mature as volcano right now. I hope we can have a batch system collaborating with the upstream smoothly, under the k/sig is a good approach.

My case is mostly about GPU sharing and network-architecture aware or stuff like that.

[Huang-Wei]

The codebase is not bound to k8s version

it does actually: https://github.com/kubernetes-sigs/kube-batch/blob/b3fc2cd0cfe5293c61c19a1134c1ea9eb526f22f/go.mod#L19-L25

[k82cn]

But it has its own ecosystem, it has its own job api, job queueing system, but we're now supporting it upstream, although not as mature as volcano right now. I hope we can have a batch system collaborating with the upstream smoothly, under the k/sig is a good approach.

The Volcano can also support customized jobs, including the job in upstream. For queueing system, kube-batch and volcano are similar; as both of them are built on my experience/understanding on LSF/Slurm (also integrate some ideas on cloud native), which is a bit different with k8s.

For the collaboration, both k8s/batch-wg and CNCF/batch-wg are the good place to discussion requirements.

[kerthcet]

Exactly, thanks @k82cn for all your efforts.

[lowang-bh]

I have developed kube-batch base on this codebase, because there are still old k8s version in used. I would like to submit those fixed issues.

[mrbobbytables]

Repo is to be archived, for more information please see: kubernetes/org#4200

/close

[k8s-ci-robot]

@mrbobbytables: Closing this issue.

In response to this:

Repo is to be archived, for more information please see: kubernetes/org#4200

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.