FAQ Click here for latest

Warning

You are currently viewing v"1.4" of the documentation and it is not the latest. For the most recent documentation, kindly click here.

What is KEDA and why is it useful?

KEDA stands for Kubernetes Event-driven Autoscaler. It is built to be able to activate a Kubernetes deployment (i.e. no pods to a single pod) and subsequently to more pods based on events from various event sources.

What are the prerequisites for using KEDA?

KEDA is designed to be run on any Kubernetes cluster. It uses a CRD (custom resource definition) and the Kubernetes metric server so you will have to use a Kubernetes version which supports these. Any Kubernetes cluster >= 1.13.0 has been tested and should work.

Does KEDA depend on any Azure service?

No, KEDA only takes a dependency on standard Kubernetes constructs and can run on any Kubernetes cluster whether in OpenShift, AKS, GKE, EKS or your own infrastructure.

Does KEDA only work with Azure Functions?

No, KEDA can scale up/down any container that you specify in your deployment. There has been work done in the Azure Functions tooling to make it easy to scale an Azure Function container.

Why should we use KEDA if we are already using Azure Functions in Azure?

There are a few reasons for this:

  • Run functions on-premises (potentially in something like an ‘intelligent edge’ architecture)
  • Run functions alongside other Kubernetes apps (maybe in a restricted network, app mesh, custom environment, etc.)
  • Run functions outside of Azure (no vendor lock-in)
  • Specific need for more control (GPU enabled compute clusters, policies, etc.)

Can I scale my HTTP container or function with KEDA and Kubernetes?

KEDA will scale a container using metrics from a scaler, but unfortunately there is no scaler today for HTTP workloads.

We recommend using the Prometheus scaler to create scale rule based on metrics around HTTP events for now. Read Anirudh Garg’s blog post to learn more.

Where can I get to the code for the Scalers?

All scalers have their code here.

Is short polling intervals a problem?

Polling interval really only impacts the time-to-activation (scaling from 0 to 1) but once scaled to one it’s really up to the HPA (horizontal pod autoscaler) which polls KEDA.

How can I get involved?

There are several ways to get involved.

  • Pick up an issue to work on. A good place to start might be issues which are marked as Good First Issue or Help Wanted
  • We are always looking to add more scalers.
  • We are always looking for more samples, documentation, etc.
  • Please join us in our weekly standup.

Can KEDA be used in production?

Yes! KEDA is now 1.0 and suited for production workloads.

What does it cost?

There is no charge for using KEDA itself.

How do I access KEDA resources using client-go?

KEDA resources can be accessed using the dynamic client from the client-go package. The dynamic client’s Resource() method accepts a GroupVersionResource describing the type of resource to be accessed and returns a NamespaceableResourceInterface which contains methods to retrieve, create, or manipulate that resource. Here’s a code sample containing a function that retrieves a KEDA ScaledObject resource by name.

package main

import (
	"fmt"
	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
	"k8s.io/apimachinery/pkg/runtime/schema"
	"k8s.io/client-go/dynamic"
	_ "k8s.io/client-go/plugin/pkg/client/auth/gcp"
	"k8s.io/client-go/tools/clientcmd"
	"os"
)

var (
	kedaGVR = schema.GroupVersionResource{
		Group:    "keda.k8s.io",
		Version:  "v1alpha1",
		Resource: "scaledobjects",
	}
)

func GetScaledObjectByName(name string) {
	config, err := clientcmd.BuildConfigFromFlags("", os.Getenv("HOME")+"/.kube/config")
	dynClient, err := dynamic.NewForConfig(config)
	if err != nil {
		panic(err)
	}
	scaledObjectClient := dynClient.Resource(kedaGVR)
	scaledObject, err := scaledObjectClient.Namespace("default").Get(name, metav1.GetOptions{})
	if err != nil {
		fmt.Printf("Error retrieving scaledobjects: %s
", err)
	} else {
		fmt.Printf("Got ScaledObject:
 %v", scaledObject)
	}
}