flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stefan Richter (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-5036) Perform the grouping of keys in restoring instead of checkpointing
Date Wed, 09 Nov 2016 09:45:59 GMT

    [ https://issues.apache.org/jira/browse/FLINK-5036?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15650479#comment-15650479
] 

Stefan Richter commented on FLINK-5036:
---------------------------------------

Ok, now I see that you are talking about something different than I assumed. I agree that
we should again provide a snapshot mechanism that uses RocksDBs native (incremental ?) snapshot
mechanics and does not iterate all K/V pairs in Flink.

> Perform the grouping of keys in restoring instead of checkpointing
> ------------------------------------------------------------------
>
>                 Key: FLINK-5036
>                 URL: https://issues.apache.org/jira/browse/FLINK-5036
>             Project: Flink
>          Issue Type: Bug
>          Components: State Backends, Checkpointing
>            Reporter: Xiaogang Shi
>
> Whenever taking snapshots of {{RocksDBKeyedStateBackend}}, the values in the states will
be written onto different files according to their key groups. The procedure is very costly
when the states are very big. 
> Given that the snapshot operations will be performed much more frequently than restoring,
we can leave the key groups as they are to improve the overall performance. In other words,
we can perform the grouping of keys in restoring instead of in checkpointing.
> I think, the implementation will be very similar to the restoring of non-partitioned
states. Each task will receive a collection of snapshots each of which contains a set of key
groups. Each task will restore its states from the given snapshots by picking values in assigned
key groups.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message