[
https://issues.apache.org/jira/browse/FLINK-14608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16969989#comment-16969989
]
Dawid Wysakowicz commented on FLINK-14608:
------------------------------------------
I am not sure about that. I fully agree that wrongly written streams will be
slower. But there are also cases when its actually the opposite (e.g. example
how important it is to write a properly optimized streams:
https://blog.overops.com/benchmark-how-java-8-lambdas-and-streams-can-make-your-code-5-times-slower/).
The differences should especially visible when there are some lazy
operations.
I believe this is a case when for-loops will perform better (I actually checked
that myself with a JMH benchmark and it is most probably true (at least for
strings as JsonNodes). I think it is good to be cautious though and make such
decisions with awareness rather than removing all streams. I'm not saying it
was not the case here. I was rather curious about the results.
> avoid using Java Streams in JsonRowDeserializationSchema
> --------------------------------------------------------
>
> Key: FLINK-14608
> URL: https://issues.apache.org/jira/browse/FLINK-14608
> Project: Flink
> Issue Type: Improvement
> Components: Formats (JSON, Avro, Parquet, ORC, SequenceFile)
> Affects Versions: 1.10.0
> Reporter: Kurt Young
> Assignee: Jiayi Liao
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.10.0
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> According to
> [https://flink.apache.org/contributing/code-style-and-quality-java.html], we
> should avoid using Java Streams in any performance critical code. Since this
> `DeserializationRuntimeConverter` will be called per field of each coming
> record, we should provide a non Java Streams implementation.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
