The Metrics Operator

Welcome to the Metrics Operator Documentation!

The Metrics Operator is a Kubernetes Cluster Operator that you can install to your cluster to easily measure different aspects of performance, including (but not limited to):

  • I/O metrics to assess storage options

  • system performance metrics (memory, cpu, etc.)

  • network or other custom metrics / timings

For this project, we aim to provide the following:

  1. A catalog of pre-defined metrics and associated containers you can quickly use

  2. A number of containerized HPC applications and experiments to demonstrate using the operator.

We wanted to create this operator because we didn’t have a solid understanding of our application performance (typically from the high performance computing space) on Kubernetes. We also didn’t have a good catalog of sample applications, and wanted to provide this to the community.

The Metrics Operator is currently 🚧️ Under Construction! 🚧️ This is a converged computing project that aims to unite the worlds and technologies typical of cloud computing and high performance computing.

To get started, check out the links below! Would you like to request a feature or contribute? Open an issue.

Last update: Nov 27, 2023