[ https://issues.apache.org/jira/browse/BEAM-8121?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16936750#comment-16936750 ]
Alexey Romanenko commented on BEAM-8121: ---------------------------------------- [~TauJan] Thank you for running these tests, I think it's very helpful. {quote} So my conclusions were that slow pipeline transformation steps decreases general throughput. So I assume its fusion issue? {quote} Yes, I think so too (as I guessed in my first comment). On the other hand, the problem with reshuffle could be serialisation/deserialisation stage (I guess CPU 100% in your case is because of this), which also can take significant amount of time. So, in general, use or not use a fusion/reshuffle - it should be a question of finding a trade-off. > Messages are not distributed per machines when consuming from Kafka topic > with 1 partition > ------------------------------------------------------------------------------------------ > > Key: BEAM-8121 > URL: https://issues.apache.org/jira/browse/BEAM-8121 > Project: Beam > Issue Type: Bug > Components: io-java-kafka > Affects Versions: 2.14.0 > Reporter: TJ > Priority: Major > Attachments: datalake-dataflow-cleaned.zip > > > Messages are consumed from Kafka using KafkaIO. Each kafka topic contains > only 1 partition. (That means that messages can be consumed only by one > Consumer per 1 consumer group) > When backlog of topic grows and system scales from 1 to X machines, all the > messages seems to be executed onĀ the same machine on which they are read. > Due to that message throughput doesn't increase comparing X machines to 1 > machine. If one machine was reading 2K messagesĀ per s, X machines will be > reading the same amount. -- This message was sent by Atlassian Jira (v8.3.4#803005)