[
https://issues.apache.org/jira/browse/KAFKA-6788?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16530449#comment-16530449
]
Colin P. McCabe commented on KAFKA-6788:
----------------------------------------
This is an optimization which we'd like to do at some point, and which hasn't
been done yet. Basically, the optimization is that if you have a bunch of
groups in the batch request which all share a common group coordinator, we'd
like to send one RPC to that group coordinator rather than several.
I think this would be a hard optimization to do correctly because of the error
handling issues. If you get an error for some, but not all, elements of the
batch, you want to retry just those elements.
The current closed PR looks like it got some wires crossed. It seems to be
replacing DescribeGroups with ListGroups, which isn't what we want here.
Listing all the groups is not very efficient.
> Grouping consumer requests per consumer coordinator in admin client
> -------------------------------------------------------------------
>
> Key: KAFKA-6788
> URL: https://issues.apache.org/jira/browse/KAFKA-6788
> Project: Kafka
> Issue Type: Improvement
> Components: admin
> Reporter: Guozhang Wang
> Assignee: Yishun Guan
> Priority: Major
> Labels: newbie++
>
> In KafkaAdminClient, for some requests like describeGroup and deleteGroup, we
> will first try to get the coordinator for each requested group id, and then
> send the corresponding request for that group id. However, different group
> ids could be hosted on the same coordinator, and these requests do support
> multi group ids be sent within the same request. So we can consider optimize
> it by grouping the requests per coordinator destination.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
