So far, we have learned what to expect from an orchestrator and the basic components required to make it work. We mentioned distributing application components on different hosts. To be able to distribute application components, we will need to deploy a cluster. A cluster is a set of nodes working together. Deploying an application to a host should be similar to deploying the same application to a cluster. The orchestrator will manage the entire workflow, and this process should be transparent for us.
Orchestrators usually manage nodes with different roles. Depending on the kind of processes those nodes run, we will define manager and worker nodes. The names may differ for each orchestrator implementation, but the logic will be the same. Manager nodes execute the orchestration control plane, while workers execute the application deployments. Worker nodes, therefore, are compute nodes.
Control plane nodes manage all the actions required for an orchestration...