A cluster is described as a series of nodes which vie for write leadership over a number of shards of the keyspace. The keyspace is divided upon cluster declaration into a predefined number of shards.
Notes:
- Nodes calculate their nodeID integer value by fetching the identities of all cluster members, lexographically sorting them (since node identities are ulids), then ascertaining the offset into array of IDs of their ID.