Today, we're excited to announce an extension of the Ternary Reporting Engine that now gives you the power to query the data sourced from your Kubernetes clusters in any way that you would like.

The data we're making available falls into two major categories:

Kubernetes Containers - CPU and Memory Request and Utilization Data from the applications and workloads you run inside your clusters. Think Pods, Deployments, ReplicaSets and DaemonSets. Use this to make sure your Kubernetes Containers are using what they're requesting and to motivate rightsizing at the application level.

Kubernetes Nodes - CPU and Memory Capacity and Utilization Data from the Nodes that make up your Kubernetes Clusters. Use this to make sure your Kubernetes Clusters are appropriately sized for the applications scheduled to them and to motivate rightsizing at the cluster level.

In each data source, cost, usage, and waste metrics are made available for you to analyze this from every perspective possible. Billing is the default data source to visualize all cloud usage and costs. To begin using Kubernetes Data in Reporting Engine, simply choose the Kubernetes Container or Kubernetes Node option in the new Data Source dropdown when creating a Report. 

Then, choose the appropriate filters, groupings and measures corresponding to what you'd like to discover. We allow you to filter and group on clusters and namespaces. Choose the measures you'd like to chart, such as cost or CPU/memory usage/wastage and you are done. In this example, I am interested in exploring container cores requested and the associated core cost for all projects (I may decide to add a filter to drill down on a few specific projects)

Save your report and add it to a dashboard and build a collection of Kubernetes reports.

You can view a complete list of our container and node metric definitions here. We hope that this feature helps you better understand your Kubernetes clusters and workloads.