Deploy the SageMath helm chart on AWS EKS

Question

Pulumi · Accepted Answer

To deploy the SageMath Helm chart on AWS EKS, we will need to follow these steps:

1. **Create an Amazon EKS Cluster**: This will be the Kubernetes cluster where our Helm chart will be deployed. We'll create a cluster with the necessary configurations using the `eks.Cluster` class.

2. **Set up a Node Group for the EKS Cluster**: For the cluster to run our pods, we need a group of worker nodes. This is done using the `aws-native.eks.Nodegroup` class.

3. **Install Helm on the Cluster**: Helm is a package manager for Kubernetes that simplifies deployment of applications. Pulumi provides the `kubernetes.helm.v3.Chart` class to deploy Helm charts.

4. **Deploy the SageMath Helm Chart**: We'll deploy the SageMath Helm chart onto our EKS cluster using the same `kubernetes.helm.v3.Chart` class, but with the chart details specific to SageMath.

Let's walk through the Pulumi TypeScript program to accomplish this:

```typescript
import * as eks from '@pulumi/eks';
import * as aws from '@pulumi/aws';
import * as k8s from '@pulumi/kubernetes';

// Create an EKS cluster.
const cluster = new eks.Cluster('my-cluster', {
    instanceType: 't2.medium', // Choose an instance type appropriate for your workload.
    desiredCapacity: 2, // Set the desired number of worker nodes.
    minSize: 1,
    maxSize: 3,
    storageClasses: 'gp2', // Define the storage class.
    deployDashboard: false, // Decide whether to deploy the Kubernetes dashboard (not recommended in production).
});

// Deploy the SageMath Helm chart into the EKS cluster.
const sageMathChart = new k8s.helm.v3.Chart('sagemath', {
    chart: 'sagemath', // The name of the Helm chart.
    version: 'your-chart-version', // Specify the version you want to deploy.
    fetchOpts: {
        repo: 'http://example-repo.com/charts', // The Helm repository URL where SageMath is hosted.
    },
    // Values to pass to the Helm chart.
    values: {
        service: {
            type: 'LoadBalancer',
        },
    },
}, { provider: cluster.provider });

// Export the cluster kubeconfig and the SageMath service endpoint.
export const kubeconfig = cluster.kubeconfig;
export const sageMathEndpoint = sageMathChart.getResourceProperty('v1/Service', 'sagemath-sagemath', 'status')
    .apply(status => status.loadBalancer.ingress[0].hostname || status.loadBalancer.ingress[0].ip);
```

Before running this Pulumi program, make sure that the Helm chart name `sagemath` and the repository `http://example-repo.com/charts` are replaced with the actual chart name and repository URL for SageMath. The chart version should also be specified according to the version you intend to deploy.

When the program is run, Pulumi will provision an EKS cluster and deploy the SageMath Helm chart onto it. The `kubeconfig` output can be used to interact with the cluster using `kubectl`, and `sageMathEndpoint` will provide the endpoint through which you can access your SageMath deployment.

This program assumes you've already set up Pulumi with the necessary AWS credentials. The `eks.Cluster` resource automates the creation of a VPC, subnets, and other necessary networking resources if they are not specified. You can customize your VPC and networking setup according to your needs.

When running the program, Pulumi performs the following actions:
- It sends the computed state of the resources to the Pulumi engine, which plans the deployment.
- If confirmed, Pulumi then executes the plan and provisions the resources in the specified cloud provider—in this case, AWS.
- Once the infrastructure is up and running, it also deploys the Helm chart, effectively starting the SageMath application in the EKS cluster.

To execute this program:
1. Save the file with a `.ts` extension (e.g., `sagemath_eks.ts`).
2. Open your terminal, navigate to the directory containing the file, and run `pulumi up`.
3. Follow the prompts from the Pulumi CLI to create the resources.