Skip to content

Commit bd64ae0

Browse files
authored
docs: updating docs to cover helm and k8s 1.33 (#140)
1 parent d3d3182 commit bd64ae0

File tree

1 file changed

+16
-25
lines changed

1 file changed

+16
-25
lines changed

README.md

Lines changed: 16 additions & 25 deletions
Original file line numberDiff line numberDiff line change
@@ -33,20 +33,22 @@ Note: this is not an officially supported Google product.
3333
The primary use cases for Kube Startup CPU Boosts are workloads that require extra CPU resources during
3434
the startup phase - typically JVM based applications.
3535

36-
The Kube Startup CPU Boost leverages [In-place Resource Resize for Kubernetes Pods](https://kubernetes.io/blog/2023/05/12/in-place-pod-resize-alpha/)
36+
The Kube Startup CPU Boost leverages [In-place Resource Resize for Kubernetes Pods](https://kubernetes.io/docs/tasks/configure-pod-container/resize-container-resources/)
3737
feature introduced in Kubernetes 1.27. It allows to revert workload's CPU resource requests and limits
3838
back to their original values without the need to recreate the Pods.
3939

4040
The increase of resources is achieved by Mutating Admission Webhook. By default, the webhook also
41-
removes CPU resource limits if present. The original resource values are set by operator after a given
42-
period of time or when the POD condition is met.
41+
removes CPU resource limits if present. The original resource values are set by tge operator after a
42+
given period of time or when the POD condition is met.
4343

4444
## Installation
4545

46-
**Requires Kubernetes 1.27 or newer with `InPlacePodVerticalScaling` feature gate
47-
enabled.**
46+
### Prerequisites
4847

49-
To install the latest release of Kube Startup CPU Boost in your cluster, run the following command:
48+
* **Requires Kubernetes 1.33 or newer**.
49+
* For older clusters (>= 1.27), enable the `InPlacePodVerticalScaling` feature gate.
50+
51+
### Install with manifests file
5052

5153
<!-- x-release-please-start-version -->
5254
```sh
@@ -73,35 +75,24 @@ kubectl kustomize | kubectl apply -f -
7375
```
7476
<!-- x-release-please-end -->
7577

76-
### Installation on Kind cluster
78+
### Install with Helm
7779

78-
You can use [KIND](https://github.com/kubernetes-sigs/kind) to get a local cluster for testing.
80+
Helm installation uses self-hosted Helm chart repo [kube-startup-cpu-boost](https://google.github.io/kube-startup-cpu-boost).
7981

8082
```sh
81-
cat <<EOF > kind-poc-cluster.yaml
82-
kind: Cluster
83-
apiVersion: kind.x-k8s.io/v1alpha4
84-
name: poc
85-
nodes:
86-
- role: control-plane
87-
- role: worker
88-
- role: worker
89-
featureGates:
90-
InPlacePodVerticalScaling: true
91-
EOF
92-
kind create cluster --config kind-poc-cluster.yaml
83+
helm repo add kube-startup-cpu-boost https://google.github.io/kube-startup-cpu-boost
84+
helm repo update
85+
helm install -n kube-startup-cpu-boost-system kube-startup-cpu-boost kube-startup-cpu-boost/kube-startup-cpu-boost
9386
```
9487

9588
### Installation on GKE cluster
9689

97-
You can use [GKE Alpha cluster](https://cloud.google.com/kubernetes-engine/docs/concepts/alpha-clusters)
98-
to run against the remote cluster.
90+
Ensure that GKE is running version 1.33 or newer by using
91+
[GKE Rapid release channel](https://cloud.google.com/kubernetes-engine/docs/concepts/release-channels).
9992

10093
```sh
10194
gcloud container clusters create poc \
102-
--enable-kubernetes-alpha \
103-
--no-enable-autorepair \
104-
--no-enable-autoupgrade \
95+
--release-channel rapid \
10596
--region europe-central2
10697
```
10798

0 commit comments

Comments
 (0)