kafka-jira mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ewen Cheslack-Postava (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (KAFKA-6626) Performance bottleneck in Kafka Connect sendRecords
Date Fri, 09 Mar 2018 17:26:00 GMT

    [ https://issues.apache.org/jira/browse/KAFKA-6626?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16393217#comment-16393217
] 

Ewen Cheslack-Postava commented on KAFKA-6626:
----------------------------------------------

It cannot be a regular map, see the comment right above that field: https://github.com/apache/kafka/blob/trunk/connect/runtime/src/main/java/org/apache/kafka/connect/runtime/WorkerSourceTask.java#L79-L80

> Performance bottleneck in Kafka Connect sendRecords
> ---------------------------------------------------
>
>                 Key: KAFKA-6626
>                 URL: https://issues.apache.org/jira/browse/KAFKA-6626
>             Project: Kafka
>          Issue Type: Bug
>    Affects Versions: 1.0.0
>            Reporter: Maciej Bryński
>            Priority: Major
>         Attachments: MapPerf.java, image-2018-03-08-08-35-19-247.png
>
>
> Kafka Connect is using IdentityHashMap for storing records.
> [https://github.com/apache/kafka/blob/trunk/connect/runtime/src/main/java/org/apache/kafka/connect/runtime/WorkerSourceTask.java#L239]
> Unfortunately this solution is very slow (2-4 times slower than normal HashMap / HashSet).
> Benchmark result (code in attachment). 
> {code:java}
> Identity 4220
> Set 2115
> Map 1941
> Fast Set 2121
> {code}
> Things are even worse when using default GC configuration 
>  (-server -XX:+UseG1GC -XX:MaxGCPauseMillis=20 -XX:InitiatingHeapOccupancyPercent=35 
-Djava.awt.headless=true)
> {code:java}
> Identity 7885
> Set 2364
> Map 1548
> Fast Set 1520
> {code}
> Java version
> {code:java}
> java version "1.8.0_152"
> Java(TM) SE Runtime Environment (build 1.8.0_152-b16)
> Java HotSpot(TM) 64-Bit Server VM (build 25.152-b16, mixed mode)
> {code}
> This problem is greatly slowing Kafka Connect.
> !image-2018-03-08-08-35-19-247.png!
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message