search

openshift / cluster-node-tuning-operator

Manage node-level tuning by orchestrating the tuned daemon.
Apache License 2.0
100 stars 104 forks source link

Is it reasonable to only set hugepages? #575

Closed cbandy closed 1 year ago

cbandy commented 1 year ago

I noticed that I could not create a PerformanceProfile dedicated to huge pages. Should it be possible to create one without specifying cpu?

apiVersion: performance.openshift.io/v2
kind: PerformanceProfile
metadata:
  name: worker-hugepages
spec:
  hugepages:
    pages:
      - { size: 2M, count: 500 }
  nodeSelector:
    node-role.kubernetes.io/worker: ''
$ oc apply --server-side -f worker-hugepages.yaml
The PerformanceProfile "worker-hugepages" is invalid: spec.cpu: Required value
jmencak commented 1 year ago

You can set huge pages via Tuned CR and kernel command-line options without specifying CPUs explicitly. The main advantage of setting huge pages via a PerformanceProfile is that you can do this per NUMA node. And yes, the PerformanceProfile CRD definition has CPU constraints. /cc @yanirq

yanirq commented 1 year ago

Most of the performance profile applications and use heavily relay on CPU definition and the split between isolated CPUs and housekeeping CPUs (reserved). For only setting hugepages you can use other options mentioned in https://github.com/openshift/cluster-node-tuning-operator/issues/575#issuecomment-1552416755

openshift-bot commented 1 year ago

Issues go stale after 90d of inactivity.

Mark the issue as fresh by commenting /remove-lifecycle stale. Stale issues rot after an additional 30d of inactivity and eventually close. Exclude this issue from closing by commenting /lifecycle frozen.

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

/lifecycle stale

cbandy commented 1 year ago

/close

openshift-ci[bot] commented 1 year ago

@cbandy: Closing this issue.

In response to [this](https://github.com/openshift/cluster-node-tuning-operator/issues/575#issuecomment-1681542288): >/close Instructions for interacting with me using PR comments are available [here](https://git.k8s.io/community/contributors/guide/pull-requests.md). If you have questions or suggestions related to my behavior, please file an issue against the [kubernetes/test-infra](https://github.com/kubernetes/test-infra/issues/new?title=Prow%20issue:) repository.