deamonset/per-node resource correction

[bmiguel-teixeira]

Which component are you using?:
vertical-pod-autoscaler

Is your feature request designed to solve a problem? If so describe the problem this feature should solve.:
Currently, the VPA targets the average utilization of resources. Even for daemon sets the same logic is applied.

Deamonsets are special in the sense that their resource consumption is usually tied to what is running on that particular node. (ex: log collectors, scanners, metrics collectors etc). This resource consumption usually varies by a large margin per-node. Some nodes may have bootstrapped a while ago and are empty-ish, while some other nodes may be under load. Some nodes (on diff pools) may be different profiles.

Describe the solution you'd like.:
When enabling VPA in a daemonset application, id like to have an option (or operation mode) to toggle VPA to calculate & update resources on a per-node basis.

Instead of averaging all the pods together, it would analyze pod by pod (one in each node for daemonset) and update those pods as needed.

Describe any alternative solutions you've considered.:
N/A

Additional context.:
Taken into account that VPA already handles evictions & mutations, I believe the work would be focused on calculating resource utilization on a per-node basis (for daemonsets). Rest of the logic would mostly stay in-place and be re-used.

Is there something similar already proposed or being discussed?

Ty

[alvaroaleman]

This seems to be a duplicate of #5928

[k8s-triage-robot]

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

This bot triages un-triaged issues 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 as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

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

/lifecycle stale

[adrianmoisey]

/area vertical-pod-autoscaler

[k8s-triage-robot]

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

This bot triages un-triaged issues 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 as fresh with /remove-lifecycle rotten
• Close this issue 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 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 with /reopen
• Mark this issue as fresh with /remove-lifecycle rotten
• Offer to help out with Issue Triage

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

/close not-planned

[k8s-ci-robot]

@k8s-triage-robot: Closing this issue, marking it as "Not Planned".

In response to this:

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

This bot triages issues 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 with /reopen
• Mark this issue as fresh with /remove-lifecycle rotten
• Offer to help out with Issue Triage

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

/close not-planned

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-sigs/prow repository.