Tasks

Step-by-step instructions for performing operations with Kubernetes.

Edit This Page

Horizontal Pod Autoscaling

This document describes the current state of Horizontal Pod Autoscaling in Kubernetes.

What is Horizontal Pod Autoscaling?

With Horizontal Pod Autoscaling, Kubernetes automatically scales the number of pods in a replication controller, deployment or replica set based on observed CPU utilization (or, with alpha support, on some other, application-provided metrics).

The Horizontal Pod Autoscaler is implemented as a Kubernetes API resource and a controller. The resource determines the behavior of the controller. The controller periodically adjusts the number of replicas in a replication controller or deployment to match the observed average CPU utilization to the target specified by user.

How does the Horizontal Pod Autoscaler work?

Horizontal Pod Autoscaler diagram

The Horizontal Pod Autoscaler is implemented as a control loop, with a period controlled by the controller manager’s --horizontal-pod-autoscaler-sync-period flag (with a default value of 30 seconds).

During each period, the controller manager queries the resource utilization against the metrics specified in each HorizontalPodAutoscaler definition. The controller manager obtains the metrics from either the resource metrics API (for per-pod resource metrics), or the custom metrics API (for all other metrics).

The HorizontalPodAutoscaler controller can fetch metrics in two different ways: direct Heapster access, and REST client access.

When using direct Heapster access, the HorizontalPodAutoscaler queries Heapster directly through the API server’s service proxy subresource. Heapster needs to be deployed on the cluster and running in the kube-system namespace.

See Support for custom metrics for more details on REST client access.

The autoscaler accesses corresponding replication controller, deployment or replica set by scale sub-resource. Scale is an interface that allows you to dynamically set the number of replicas and examine each of their current states. More details on scale sub-resource can be found here.

API Object

The Horizontal Pod Autoscaler is an API resource in the Kubernetes autoscaling API group. The current stable version, which only includes support for CPU autoscaling, can be found in the autoscaling/v1 API version.

The alpha version, which includes support for scaling on memory and custom metrics, can be found in autoscaling/v2alpha1. The new fields introduced in autoscaling/v2alpha1 are preserved as annotations when working with autoscaling/v1.

More details about the API object can be found at HorizontalPodAutoscaler Object.

Support for Horizontal Pod Autoscaler in kubectl

Horizontal Pod Autoscaler, like every API resource, is supported in a standard way by kubectl. We can create a new autoscaler using kubectl create command. We can list autoscalers by kubectl get hpa and get detailed description by kubectl describe hpa. Finally, we can delete an autoscaler using kubectl delete hpa.

In addition, there is a special kubectl autoscale command for easy creation of a Horizontal Pod Autoscaler. For instance, executing kubectl autoscale rc foo --min=2 --max=5 --cpu-percent=80 will create an autoscaler for replication controller foo, with target CPU utilization set to 80% and the number of replicas between 2 and 5. The detailed documentation of kubectl autoscale can be found here.

Autoscaling during rolling update

Currently in Kubernetes, it is possible to perform a rolling update by managing replication controllers directly, or by using the deployment object, which manages the underlying replication controllers for you. Horizontal Pod Autoscaler only supports the latter approach: the Horizontal Pod Autoscaler is bound to the deployment object, it sets the size for the deployment object, and the deployment is responsible for setting sizes of underlying replication controllers.

Horizontal Pod Autoscaler does not work with rolling update using direct manipulation of replication controllers, i.e. you cannot bind a Horizontal Pod Autoscaler to a replication controller and do rolling update (e.g. using kubectl rolling-update). The reason this doesn’t work is that when rolling update creates a new replication controller, the Horizontal Pod Autoscaler will not be bound to the new replication controller.

Support for multiple metrics

Kubernetes 1.6 adds support for scaling based on multiple metrics. You can use the autoscaling/v2alpha1 API version to specify multiple metrics for the Horizontal Pod Autoscaler to scale on. Then, the Horizontal Pod Autoscaler controller will evaluate each metric, and propose a new scale based on that metric. The largest of the proposed scales will be used as the new scale.

Support for custom metrics

Note: Kubernetes 1.2 added alpha support for scaling based on application-specific metrics using special annotations. Support for these annotations was removed in Kubernetes 1.6 in favor of the autoscaling/v2alpha1 API. While the old method for collecting custom metrics is still available, these metrics will not be available for use by the Horizontal Pod Autoscaler, and the former annotations for specifying which custom metrics to scale on are no longer honored by the Horizontal Pod Autoscaler controller.

Kubernetes 1.6 adds support for making use of custom metrics in the Horizontal Pod Autoscaler. You can add custom metrics for the Horizontal Pod Autoscaler to use in the autoscaling/v2alpha1 API. Kubernetes then queries the new custom metrics API to fetch the values of the appropriate custom metrics.

Prerequisites

In order to use custom metrics in the Horizontal Pod Autoscaler, you must deploy your cluster with the --horizontal-pod-autoscaler-use-rest-clients flag on the controller manager set to true. You must then configure your controller manager to speak to the API server through the API server aggregator, by setting the controller manager’s target API server to the API server aggregator (using the --apiserver flag). The resource metrics API and custom metrics API must also be registered with the API server aggregator, and must be served by API servers running on the cluster.

You can use Heapster’s implementation of the resource metrics API by running Heapster with the--api-server flag set to true. A separate component must provide the custom metrics API (more information on the custom metrics API is available at the k8s.io/metrics repository).

Further reading

Analytics

Create an Issue Edit this Page