Skip to content

This repo contains scripts to deploy a Ray cluster on a private Azure Kubernetes Service (AKS) cluster using the KubeRay operator and use it to train an AI/ML model.

License

Notifications You must be signed in to change notification settings

Azure-Samples/aks-ray-sample

Repository files navigation

aks-ray-sample

This sample shows how to deploy a Ray cluster on an Azure Kubernetes Service (AKS) cluster using the KubeRay operator. The sample also shows how to run a simple Ray application on the cluster.

Getting Started

Clone the repository to your local machine. Then make sure you have all the prerequisites installed.

Prerequisites

  1. An Azure subscription. If you don't have an Azure subscription, you can create a free account here.
  2. The Azure CLI installed on your local machine. You can install the Azure CLI by following the instructions here.
  3. The Azure Kubernetes Service Preview extension must be installed.
  4. Helm must be installed.
  5. Terraform client tools or OpenTofu must be installed. This guide makes use of Terrafrom, however the modules used should be compatible with OpenTofu.

Quickstart

  1. Clone the repository https://github.com/Azure-Samples/aks-ray-sample on to your local machine.
  2. Navigate to the directory you cloned the repository to.
  3. Enable execute permissions on the deploy.sh script by running chmod +x deploy.sh.
  4. Run the deploy.sh script by running ./deploy.sh. This script will deploy the AKS cluster, install the KubeRay operator then submit a training job to run on the AKS cluster.

Resources

About

This repo contains scripts to deploy a Ray cluster on a private Azure Kubernetes Service (AKS) cluster using the KubeRay operator and use it to train an AI/ML model.

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •