A self-service cluster is a Kubernetes cluster that your organization runs and you reuse to run mappings.
The Kubernetes cluster can run on AWS or Microsoft Azure. You can use self-managed clusters, including Amazon EC2 instances and Azure Virtual Machines, or you can use service-managed clusters, including Amazon Elastic Kubernetes Service and Azure Kubernetes Service.
Using a self-service cluster gives you finer controls over the compute environment by providing isolation through namespaces, contexts, annotations, and tolerations. Because you manage the cluster, the Secure Agent needs minimal permissions in your environment.
Compared to a fully-managed cluster, a self-service cluster provides the following advantages:
•You have more control over the cluster control plane.
•You have full access to the cluster and manage all components.
•You have more control over the deployment and administration of your cluster. For example, you can implement multiple node groups or use different instance types for different nodes.
To connect the Secure Agent to a self-service cluster, use the kubeconfig file that is generated for the cluster. The kubeconfig file is a YAML file that contains the cluster configuration. Enter the path to the kubeconfig file in the advanced configuration so that the Secure Agent can connect to the self-service cluster and submit jobs to the cluster.
Because the Secure Agent doesn’t manage the cluster, the Secure Agent doesn’t scale the cluster based on the workload. When you shut down the cluster, the Secure Agent removes all job-related resources from the cluster.