[
https://issues.apache.org/jira/browse/KAFKA-16205?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
David Arthur updated KAFKA-16205:
---------------------------------
Component/s: (was: controller)
> Reduce number of metadata requests during hybrid mode
> -----------------------------------------------------
>
> Key: KAFKA-16205
> URL: https://issues.apache.org/jira/browse/KAFKA-16205
> Project: Kafka
> Issue Type: Improvement
> Components: kraft, migration
> Affects Versions: 3.4.0, 3.5.0, 3.6.0, 3.7.0
> Reporter: David Arthur
> Priority: Major
>
> When migrating a cluster with a high number of brokers and partitions, it is
> possible for the controller channel manager queue to get backed up. This can
> happen when many small RPCs are generated in response to several small
> MetadataDeltas being handled MigrationPropagator.
>
> In the ZK controller, various optimizations have been made over the years to
> reduce the number of UMR and LISR sent during controlled shutdown or other
> large metadata events. For the ZK to KRaft migration, we use the
> MetadataLoader infrastructure to learn about and propagate metadata to ZK
> brokers.
>
> We need to improve the batching in MigrationPropagator to avoid performance
> issues during the migration of large clusters.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
