StorageOS clusters represent groups of nodes which run a common distributed control plane.

Typically, a StorageOS cluster maps one-to-one to a Kubernetes (or similar orchestrator) cluster, and we expect our daemonset to run on all worker nodes within the cluster that will consume or present storage.

Clusters use etcd to maintain state and manage distributed consensus between nodes.