[
https://issues.apache.org/jira/browse/KAFKA-7149?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16544337#comment-16544337
]
Guozhang Wang commented on KAFKA-7149:
--------------------------------------
Searched on the JIRA list but did not find it. Maybe we once discussed about it
on the mailing list, but not created a ticket for it yet? As we may likely add
more information with the metadata version bumped up to 4:
1) KAFKA-5037: added an global error code ([~yuzhih...@gmail.com])
2) KAFKA-4696: state-ware standby task assignment ([~Yohan123])
3) KAFKA-6718: rack-aware task assignment ([~_deepakgoyal])
4) KIP-258: state store format change with embedded timestamp ([~mjsax])
I think it urges us to consider this issue sooner than later. So I'd like to
collect more feedbacks on the tentative proposals, also cc'ed all the
contributors of the above features to think about how to reduce the encoded
bytes.
> Reduce assignment data size to improve kafka streams scalability
> ----------------------------------------------------------------
>
> Key: KAFKA-7149
> URL: https://issues.apache.org/jira/browse/KAFKA-7149
> Project: Kafka
> Issue Type: Improvement
> Components: streams
> Reporter: Ashish Surana
> Assignee: Ashish Surana
> Priority: Major
>
> We observed that when we have high number of partitions, instances or
> stream-threads, assignment-data size grows too fast and we start getting
> below RecordTooLargeException at kafka-broker.
> Workaround of this issue is commented at:
> https://issues.apache.org/jira/browse/KAFKA-6976
> Still it limits the scalability of kafka streams as moving around 100MBs of
> assignment data for each rebalancing affects performance & reliability
> (timeout exceptions starts appearing) as well. Also this limits kafka streams
> scale even with high max.message.bytes setting as data size increases pretty
> quickly with number of partitions, instances or stream-threads.
>
> Solution:
> To address this issue in our cluster, we are sending the compressed
> assignment-data. We saw assignment-data size reduced by 8X-10X. This improved
> the kafka streams scalability drastically for us and we could now run it with
> more than 8,000 partitions.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
