You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kafka.apache.org by "ASF GitHub Bot (JIRA)" <ji...@apache.org> on 2016/11/30 02:32:58 UTC
[jira] [Commented] (KAFKA-4469) Consumer throughput regression
caused by decrease in max.poll.records
[ https://issues.apache.org/jira/browse/KAFKA-4469?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15707289#comment-15707289 ]
ASF GitHub Bot commented on KAFKA-4469:
---------------------------------------
GitHub user hachikuji opened a pull request:
https://github.com/apache/kafka/pull/2190
KAFKA-4469: Fix consumer performance regression from unneeded list copy
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/hachikuji/kafka KAFKA-4469
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/kafka/pull/2190.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #2190
----
commit 701073de6163e66bac45ebb12130f3789f507531
Author: Jason Gustafson <ja...@confluent.io>
Date: 2016-11-30T01:50:32Z
KAFKA-4469: Fix consumer performance regression from unneeded list copy
----
> Consumer throughput regression caused by decrease in max.poll.records
> ---------------------------------------------------------------------
>
> Key: KAFKA-4469
> URL: https://issues.apache.org/jira/browse/KAFKA-4469
> Project: Kafka
> Issue Type: Bug
> Affects Versions: 0.10.1.0
> Reporter: Jason Gustafson
> Assignee: Jason Gustafson
> Fix For: 0.10.1.1
>
>
> There appears to be a small performance regression in 0.10.1.0 from previous versions. I tracked it back to KAFKA-3888. As part of KIP-62, we decreased the value of {{max.poll.records}} from {{Integer.MAX_VALUE}} to 500. Based on some performance testing, this results in about a 5% decrease in throughput. This depends on the fetch and message sizes. My test used message size of 1K with the default fetch size, and the default {{max.poll.records}} of 500.
> The main cause of the regression seems to be an unneeded list copy in {{Fetcher}}. Basically when we have more records than we need to satisfy {{max.poll.records}}, then we copy the fetched records into a new list. When I modified the code to use a sub-list, which does not need a copy, the performance is much closer to that of 0.10.0 (within 1% or so with lots of qualification since there are many unexplored parameters). The remaining performance gap could be explained by sub-optimal pipelining as a result of KAFKA-4007 (this is likely part of the story anyway based on some rough testing).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)