Sihua Zhou created FLINK-8602:

             Summary: Accelerate recover from failover when use incremental 
                 Key: FLINK-8602
             Project: Flink
          Issue Type: Improvement
          Components: State Backends, Checkpointing
    Affects Versions: 1.4.0
            Reporter: Sihua Zhou
            Assignee: Sihua Zhou

Currently, when enable incremental checkpoint, if user change the parallelism 
then `hasExtraKeys` may be `true`. If this occur, flink will loop all rocksdb 
instance and iterator all data to fetch the data that fails into current 
`KeyGroupRange`, this can be improved because

 if a state handle's `KeyGroupRange` is fully covered by  current 
`KeyGroupRange`, we can open the rocksdb it corresponded directly.


This message was sent by Atlassian JIRA

Reply via email to