Skip to content

Latest commit

Β 

History

History
30 lines (23 loc) Β· 1.41 KB

RESOURCES.md

File metadata and controls

30 lines (23 loc) Β· 1.41 KB

Resources and Links

Large Clusters and Scaling

  • Scaling Kubernetes to 2,500 Nodes - A blog post from the OpenAI team on some of the issues and best practices associated with running large scale Kubernetes clusters.

Software, Frameworks and Collections

Software

  • Volcano - A Kubernetes native batch scheduler. Adds support for MPI, fair-share, queues and more.
  • Kubeflow - Tightly integrated collection of "best of breed" software for Machine learning.
  • Zero to JupterHub - Helm Chart for Jupyterhub maintained by upstream Jupyterhub community.
  • Armada - A multi-cluster batch scheduler for high-throughput workloads on Kubernetes

Frameworks

Research Institution Collections