Analytics dashboard for Kubernetes

go get



Performance analytics UI for Kubernetes Clusters.

The goal of Kubedash is to allow the user or an administrator of a Kubernetes cluster to easily verify and understand the performance of a cluster and jobs running within it through intuitive visualizations of aggregated metrics, derived stats and event patterns. It is not intended to be a general-purpose Kubernetes UI. Instead, kubedash uses multiple sources of information to summarize and provide high-level analytic information to users and to the cluster administrator.

Some of the current features are:

  • Resource utilization views of Cluster, Nodes, Namespaces, Pods and Containers for the past 1 hour.
  • Derived stats for each resource over the past 1 day.
  • Indications of containers with no resource limits.

Future features include the following:

  • Recommendations of resource limits for individual containers.
  • Detecting and highlighting resource starvation in a cluster.
  • Detecting and highlighting crash-looping and misbehaving containers.
  • Support for other cluster management systems like CoreOS, Swarm, etc.


To use Kubedash, the following are required:

  • A Kubernetes cluster of v1.0 or higher is operational and accessible from kubectl.
  • Heapster v0.18.0 is running as the heapster service on the kube-system namespace.

After cloning this repository, use the following command to create the Kubedash pod:

kubectl create -f deploy/kube-config.yaml

To access the Kubedash UI, visit the following URL: https://<kubernetes-master>/api/v1/proxy/namespaces/kube-system/services/kubedash/ where <kubernetes-master> is the IP address of the kubernetes master node.


Kubedash is based on three components-

Heapster provides the data for Kubedash. Heapster runs as a service by default in all kubernetes clusters, collecting metrics and analytics for individual containers. The Heapster Model exposes aggregated metrics and statistics that are relevant for cluster-level analytics through a RESTful API.

Kubedash provides the other two pieces - a web server relaying REST calls, managing sockets and providing additional authentication and, a frontend to provide visualizations for the aggregated metrics and statistics of interest.


The work done has been licensed under Apache License 2.0.The license file can be found here. You can find out more about license,at