Does GKE Autopilot manages autoscaling by default?

user_11450710 0
Viewed 176

Good morning. I am doing some tests with the new Google Cloud Kubernetes Engine's Autopilot mode. I know that it automates a lot of the machine resources' management, but I am not sure about what it automates. Does it only cares about provisioning the hardware resources that I set inside my PodSpec? Or does it also cares about scaling up and down the number of containers that I have based on traffic intensity?

I am coming from Cloud Run, so, after all, my main question is: Now, with GKE Autopilot, do I need to do something for it to create new container instances when the traffic intensity increases or is it all automatically managed? Do I need to set HPA, VPA and other autoscaler technologies when using autopilot?

google-kubernetes-engine google-cloud-platform
1 Answers

For GKE autopilot you need to create the HPA and VPA configuration

GKE autopilot will the scaling of Node by default

You can read more at : https://cloud.google.com/kubernetes-engine/docs/concepts/autopilot-overview#comparison

Scaling Pre-configured: Autopilot handles all the scaling and configuring of your nodes.

Default: You configure Horizontal pod autoscaling (HPA) You configure Vertical Pod autoscaling (VPA)

Do I need to set HPA, VPA and other autoscaler technologies when using autopilot?

Autoscaler is not required as it will be by default managed by GKE and will scale the Node as per requirement.

user_55258240
Related