You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@beam.apache.org by "Soumabrata Chakraborty (Jira)" <ji...@apache.org> on 2020/03/05 14:21:00 UTC
[jira] [Commented] (BEAM-9440) Performance Issue with Spark Runner
compared with Native Spark
[ https://issues.apache.org/jira/browse/BEAM-9440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17052177#comment-17052177 ]
Soumabrata Chakraborty commented on BEAM-9440:
----------------------------------------------
[~iemejia] - Its not just a performance issue of the Spark Runner alone.
Based on the article I cited in the description – even other Runners like Flink, etc display similar detoriation in performance as opposed to writing software in the original framework.
So, Spark Runner on Beam is Slower than Spark by a large margin. Similarly, Flink Runner on Beam is Slower than Flink by a big margin – and so on.
> Performance Issue with Spark Runner compared with Native Spark
> --------------------------------------------------------------
>
> Key: BEAM-9440
> URL: https://issues.apache.org/jira/browse/BEAM-9440
> Project: Beam
> Issue Type: Bug
> Components: runner-spark
> Reporter: Soumabrata Chakraborty
> Priority: Major
>
> While doing a performance evaluation of Apache Beam with Spark Runner - I found that even for a simple word count problem on a text file – Beam with Spark runner was slower by a factor of 5 times as compared to Spark for a dataset as small as 14 GB.
> You will find more details on this evaluation here - [https://github.com/soumabrata-chakraborty/spark-vs-beam/blob/master/README.md]
> I also came across this analysis called _**Quantitative Impact Evaluation of an Abstraction Layer for Data Stream Processing Systems_ ([https://arxiv.org/pdf/1907.08302.pdf])
> According to it, the observation was that for most scenarios the slowdown was at least a factor of 3 with the worse case being a factor of 58!
> While it is understood that an abstraction layer would come with some performance cost - the current performance cost seems to be very high.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)