[jira] [Comment Edited] (KAFKA-3184) Add Checkpoint for In-memory State Store

["Nikolay Izhikov (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/KAFKA-3184?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17099287#comment-17099287 ] 

Nikolay Izhikov edited comment on KAFKA-3184 at 5/4/20, 8:15 PM:
-----------------------------------------------------------------

[~guozhang]

> we need to make the checkpointed offsets value to be consistent with the checkpoint image on disk itself

Can you, please, clarify this requirement.
 Right now, I call {{FileChannel.force}} on each checkpoint.
 So all data are forced to the disk.
 Is it enough?


was (Author: nizhikov):
[~guozhang]

> we need to make the checkpointed offsets value to be consistent with the checkpoint image on disk itself

Can you, please, clarify this requirement.
Right now, I call {{FileChannel.force}} on each checkpoint.
So all data are forsed to the disk.
Is it enough? 


> Add Checkpoint for In-memory State Store
> ----------------------------------------
>
>                 Key: KAFKA-3184
>                 URL: https://issues.apache.org/jira/browse/KAFKA-3184
>             Project: Kafka
>          Issue Type: Improvement
>          Components: streams
>            Reporter: Guozhang Wang
>            Assignee: Nikolay Izhikov
>            Priority: Major
>              Labels: user-experience
>
> Currently Kafka Streams does not make a checkpoint of the persistent state store upon committing, which would be expensive since it is "stopping the world" and write on disks: for example, RocksDB would require you to copy the file directory to make a copy naively. 
> However, for in-memory stores checkpointing maybe doable in an asynchronous manner hence it can be done quickly. And the benefit of having intermediate checkpoint is to avoid restoring from scratch if standby tasks are not present.



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