How to Add Node Pools to a Cluster

DigitalOcean Kubernetes (DOKS) is a managed Kubernetes service. Deploy Kubernetes clusters with a fully managed control plane, high availability, autoscaling, and native integration with DigitalOcean Load Balancers and volumes. You can add node pools using shared and dedicated CPUs, and NVIDIA H100 GPUs in a single GPU or 8 GPU configuration. DOKS clusters are compatible with standard Kubernetes toolchains and the DigitalOcean API and CLI.

Add a Node Pool to a Cluster Using Automation

Add a GPU Worker Node to an Existing Cluster

You can also add a GPU node pool to an existing cluster on versions 1.30.4-do.0, 1.29.8-do.0, 1.28.13-do.0, and later.

Note

In rare cases, it can take several hours for a GPU Droplet to provision. If you have an unusually long creation time, open a support ticket.

doctl kubernetes cluster node-pool create gpu-cluster --name gpu-worker-pool-1 --size gpu-h100x1-80gb --count 4

curl --location --request POST 'https://api.digitalocean.com/v2/kubernetes/clusters/{cluster_id}/node_pools' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer $DIGITALOCEAN_TOKEN' \
--data '{
    "node_pools": [
        {
            "size": "gpu-h100x1-80gb",
            "count": 4,
            "name": "new-gpu-worker-pool"
        }
}',

To run GPU workloads after you create a cluster, use the GPU nodes-specific labels and taint in your workload specifications to schedule pods that match. You can use a configuration spec, similar to the pod spec shown below, for your actual workloads:

    
        
            
apiVersion: v1
kind: Pod
metadata:
  name: gpu-workload
spec:
  restartPolicy: Never
  nodeSelector:
    doks.digitalocean.com/gpu-brand: nvidia
  containers:
    - name: cuda-container
      image: nvcr.io/nvidia/k8s/cuda-sample:vectoradd-cuda10.2
  tolerations:
    - key: nvidia.com/gpu
      operator: Exists
      effect: NoSchedule

        
    

The above spec shows how to create a pod that runs NVIDIA’s CUDA image and uses the labels and taint for GPU worker nodes.

You can use the cluster autoscaler to scale the GPU node pool down to 1 or use the DigitalOcean CLI or API to manually scale the node pool down to 0. For example:

  doctl kubernetes cluster node-pool update <your-cluster-id> <your-nodepool-id> --count 0 

Add a Node Pool to a Cluster Using the Control Panel

To add additional node pools to an existing cluster:

1. Open the cluster’s More menu and select View Nodes.
2. Click Add Node Pool.
3. Select the name, type, size, and number of Droplets in the pool. Click Add Node Pool(s) to add additional node pools.
4. Click Save to save your changes and provision your new nodes.