eks-blueprints-cdk-kubeflow-ext v0.1.9

eks-blueprints-cdk-kubeflow-extension

This module is a kubeflow extension for https://github.com/aws-quickstart/cdk-eks-blueprints.

Kubeflow on EKS

The Kubeflow project is dedicated to making deployments of machine learning (ML) workflows on Kubernetes simple, portable and scalable. Our goal is not to recreate other services, but to provide a straightforward way to deploy best-of-breed open-source systems for ML to diverse infrastructures. Anywhere you are running Kubernetes, you should be able to run Kubeflow.

This example deploys the following resources

• Creates EKS Cluster Control plane with public endpoint (for demo purpose only) with a managed node group
• Deploys supporting add-ons: ClusterAutoScaler, AwsLoadBalancerController, VpcCni, CoreDns, KubeProxy, EbsCsiDriver
• Deploy Kubeflow on the EKS cluster

Note: we use EKS 1.21 here which is the latest EKS version supported by Kubeflow. see reference below https://awslabs.github.io/kubeflow-manifests/docs/about/eks-compatibility/

Prerequisites:

Ensure that you have installed the following tools on your machine.

1. aws cli
2. kubectl
3. cdk
4. npm

Deploy EKS Cluster with Amazon EKS Blueprints for CDK

Clone the repository

git clone https://github.com/aws-samples/cdk-eks-blueprints-patterns.git

Create a CDK project, Bootstrap your environment and install dependency

cdk init app --language typescript
cdk bootstrap aws://<AWS_ACCOUNT_ID>/<AWS_REGION>
npm i @aws-quickstart/eks-blueprints
npm i eks-blueprints-cdk-kubeflow-ext

Replace the contents of bin/.ts with the following:

import 'source-map-support/register';
import * as cdk from 'aws-cdk-lib';
import * as blueprints from '@aws-quickstart/eks-blueprints';
import * as KubeflowAddOn from 'eks-blueprints-cdk-kubeflow-ext';
const app = new cdk.App();
// AddOns for the cluster.
const addOns: Array<blueprints.ClusterAddOn> = [
    new blueprints.addons.MetricsServerAddOn,
    new blueprints.addons.ClusterAutoScalerAddOn,
    new blueprints.addons.AwsLoadBalancerControllerAddOn(),
    new blueprints.addons.VpcCniAddOn(),
    new blueprints.addons.CoreDnsAddOn(),
    new blueprints.addons.KubeProxyAddOn(),
    new blueprints.addons.EbsCsiDriverAddOn(),
    new KubeflowAddOn({
         namespace: 'kubeflow-pipelines'
     })
];
const account = 'XXXXXXXXXXXXX'
const region = 'your region'
const props = { env: { account, region } }
new blueprints.EksBlueprint(app, { id: 'kubeflow-eks', addOns}, props)

Deploy the stack using the following command

cdk deploy

Verify the resources

Let’s verify the resources created by Steps above.

kubectl get nodes # Output shows the EKS Managed Node group nodes

kubectl get ns | kubeflow # Output shows kubeflow namespace

kubectl get pods --namespace=kubeflow-pipelines # Output shows kubeflow pods

Execute Machine learning jobs on Kubeflow

log into Kubeflow pipeline UI by creating a port-forward to the ml-pipeline-ui service

kubectl port-forward svc/ml-pipeline-ui 9000:80 -n =kubeflow-pipelines

and open this browser: http://localhost:9000/#/pipelines more pipeline examples can be found at https://www.kubeflow.org/docs/components/pipelines/tutorials/

Cleanup

To clean up your EKS Blueprints, run the following commands:

cdk destroy --all