You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@flink.apache.org by "Yun Tang (Jira)" <ji...@apache.org> on 2021/03/11 16:17:00 UTC

[jira] [Created] (FLINK-21736) Introduce latency tracking state

Yun Tang created FLINK-21736:
--------------------------------

             Summary: Introduce latency tracking state
                 Key: FLINK-21736
                 URL: https://issues.apache.org/jira/browse/FLINK-21736
             Project: Flink
          Issue Type: New Feature
          Components: Runtime / State Backends
            Reporter: Yun Tang
            Assignee: Yun Tang
             Fix For: 1.13.0


This ticket is extended from FLINK-16444 which wants to track latency of RocksDB options. However previous ticket has two problems:
 # Cannot apply to heap based state backend and other user customized state backend.
 # User need background knowledge to know what RocksDB operation corresponding to what state operation, e.g RocksDB adopts {{merge}} to act as listState#add.

Thus, I prefer to introduce latency tracking state wrapper to detect the access latency. And to not impact the original state access, I would still use sample interval as previous FLINK-16444 did.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)