This is a Rebalancer specific to custom mode. It is tasked with checking an existing mapping of
partitions against the set of live instances to mark assignment states as dropped or erroneous
The input is the required current assignment of partitions to instances, as well as the required
existing instance preferences.
The output is a verified mapping based on that preference list, i.e. partition p has a replica
on node k with state s, where s may be a dropped or error state if necessary.
Given an ideal state for a resource and the liveness of instances, compute the best possible
state assignment for each partition's replicas.
||Provides the current state and pending state transitions for all partitions
This method provides all the relevant information needed to rebalance a resource.
If you need additional information use manager.getAccessor to read the cluster data.
This allows one to compute the newIdealState according to app specific requirement.
||Name of the resource to be rebalanced
||Provides the current state and pending state transition for all
||Provides additional methods to retrieve cluster data.