[
https://issues.apache.org/jira/browse/KAFKA-8215?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16814992#comment-16814992
]
ASF GitHub Bot commented on KAFKA-8215:
---------------------------------------
ableegoldman commented on pull request #6560: KAFKA-8215: Pt I. Share block
cache between instances
URL: https://github.com/apache/kafka/pull/6560
As of v5.12, RocksDB provides an API for sharing the block cache across all
instances in a single process. Because it involves passing the same cache to
the Rocks Options, this ability is not supported by the current Streams configs.
### Committer Checklist (excluded from commit message)
- [ ] Verify design and implementation
- [ ] Verify test coverage and CI build status
- [ ] Verify documentation (including upgrade notes)
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> Limit memory usage of RocksDB
> -----------------------------
>
> Key: KAFKA-8215
> URL: https://issues.apache.org/jira/browse/KAFKA-8215
> Project: Kafka
> Issue Type: Improvement
> Components: streams
> Reporter: Sophie Blee-Goldman
> Priority: Major
>
> The memory usage of Streams is currently unbounded in part because of
> RocksDB, which consumes memory on a per-instance basis. Each instance (ie
> each persistent state store) will have its own write buffer, index blocks,
> and block cache. The size of these can be configured individually, but there
> is currently no way for a Streams app to limit the total memory available
> across instances.
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
