Create a Workload Cluster

Provision workload clusters for deploying applications and hosting environments.

Summary

Workload clusters run your applications and environments. Once you have a management cluster provisioned, you can create multiple workload clusters to organize your infrastructure by environment, team, or purpose.

Prerequisites

• Platform Admin or Team Admin role
• A management cluster provisioned for your organization
• A cloud account configured in your organization

Clusters Page Overview

After your management cluster is provisioned, the Clusters page shows a graph view of your cluster topology with the management cluster and any existing workload clusters.

From this page, click Add workload cluster to create a new cluster.

Create Workload Cluster

1. Navigate to Clusters in the left sidebar

2. Click Add workload cluster

3. The cluster creation form opens in a side panel:

Cluster Type

Choose the type of cluster to create:

• Physical: Full Kubernetes cluster with dedicated EC2 infrastructure
• Virtual: Lightweight virtual cluster that shares resources with other clusters (faster to create, lower cost)

Cluster Configuration

4. Configure the cluster settings:

   Field	Description	Example
   Cluster name	Unique name for the cluster	development
   Cloud account	AWS account to provision the cluster in	development
   Cloud region	AWS region for the cluster	us-west-2
   AMI type	Amazon Machine Image type	AL2_x86_64
   Instance size	EC2 instance type for nodes	m5.xlarge
   Number of nodes	Initial node count (minimum 2 recommended)	2

5. Optionally select an environment for the cluster to host

Example Configuration

Here's an example of a filled form for a development cluster:

6. Click Add cluster to start provisioning

Monitor Provisioning

After clicking Add cluster, the form closes and the Clusters page shows your new cluster in a provisioning state:

Workload cluster provisioning typically takes 15-20 minutes. You can:

• View provisioning progress in the cluster details panel
• Check the Logs tab for real-time provisioning output
• Monitor the cluster status in ArgoCD

Instance Size Guidelines

Use Case	Recommended Instance	Notes
Development	m5.xlarge	Good balance for dev workloads
Staging	m5.xlarge - m5.2xlarge	Match production sizing
Production	m5.2xlarge or larger	Size based on workload requirements

Important Notes

• Physical clusters require more resources and take longer to provision than virtual clusters
• Cluster names must be unique within your organization
• You cannot change the cluster type after creation
• The minimum recommended node count is 2 for high availability

What's Next?

After your workload cluster is provisioned:

• Register applications to deploy to your cluster
• Create environments to organize your deployments
• Set up pipeline templates for automated deployments