Demo - AI Workloads on OpenShift

This repo is intented to demonstrate the capibilities for OpenShift for running AI workloads based on the docs OpenShift - Ai Workloads

Prerequisites - Get a cluster

• OpenShift 4.14+
  • role: cluster-admin - for all demo or cluster configs
  • role: self-provisioner - for namespaced components

Red Hat Demo Platform Options (Tested)

NOTE: The node sizes below are the recommended minimum to select for provisioning

• AWS with OpenShift Open Environment
  • 1 x Control Plane - m6a.2xlarge
  • 0 x Workers - m6a.2xlarge
• One Node OpenShift
  • 1 x Control Plane - m6a.2xlarge
• OpenShift GitOps Blank Environment
  • 1 x Control Plane - 16 cores, 64Gi

Getting Started

Install the OpenShift Web Terminal

The following icon should appear in the top right of the OpenShift web console after you have installed the operator. Clicking this icon launches the web terminal.

NOTE: Reload the page in your browser if you do not see the icon after installing the operator.

Make the enhanced web terminal permanent

Important

Run the following commands from the enhanced web terminal

# apply the enhanced web terminal
until oc apply -k https://github.com/redhat-na-ssa/demo-ocp-ai-workloads/demo/web-terminal; do : ; done

# reset the web terminal (from the web terminal)
$(wtoctl | grep 'oc delete')

Setup at least 1 worker and isolate the control plane

ocp_machineset_scale 1
ocp_control_nodes_not_schedulable
oc apply -k ../demo_ops/components/cluster-configs/autoscale/overlays/default

Setup Nvidia GPU Node

# setup gpu node
ocp_aws_machineset_create_gpu
ocp_machineset_scale 1

Demo Quickstart

until oc apply -k demo; do : ; done

Additional Info

• See NOTES

• See KUEUE

• Kueue Docs - Upstream

• OpenShift - Ai Workloads