You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Stefan Richter (JIRA)" <ji...@apache.org> on 2018/06/01 14:25:00 UTC
[jira] [Updated] (FLINK-9491) Implement timer data structure based
on RocksDB
[ https://issues.apache.org/jira/browse/FLINK-9491?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Stefan Richter updated FLINK-9491:
----------------------------------
Summary: Implement timer data structure based on RocksDB (was: Implement timer service based on RocksDB)
> Implement timer data structure based on RocksDB
> -----------------------------------------------
>
> Key: FLINK-9491
> URL: https://issues.apache.org/jira/browse/FLINK-9491
> Project: Flink
> Issue Type: Sub-task
> Components: State Backends, Checkpointing
> Reporter: Stefan Richter
> Assignee: Stefan Richter
> Priority: Major
> Fix For: 1.6.0
>
>
> We can now implement timer state that is stored in RocksDB for users that run the {{RocksDBKeyedStateBackend}}. As explained in the design document (https://docs.google.com/document/d/1XbhJRbig5c5Ftd77d0mKND1bePyTC26Pz04EvxdA7Jc/edit#heading=h.17v0k3363r6q) this should also give us asynchronous and incremental snapshots for timer state that is larger than main memory.
> We need to think about a way in which to user can select either to run timers on RocksDB or on the heap when using the {{RocksDBKeyedStateBackend}}.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)