[
https://issues.apache.org/jira/browse/SPARK-33772?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17415191#comment-17415191
]
Dongjoon Hyun commented on SPARK-33772:
---------------------------------------
To [~medb]. There is no way to make it work smoothly. This Umbrella JIRA issue
has only a subset of issues because the community didn't start testing Java 17
seriously. I'm expecting a bunch of issues like SPARK-24417 (Java 11) work.
[~h-vetinari]. I don't think that's a release blocker for Apache Spark. Apache
Spark has its own independent release cycle.
The Apache Spark community usually work in this way.
1. Add a CI for build and fix compilation issues
2. Make UT pass module by module
3. Add a CI for run all UTs
4. Run all micro-benchmarks to check the performance regression (You can see
that our repo have both JDK8/JDK11 benchmark result)
5. Run big benchmark like TPCDS
In short, the Apache Spark repository is not ready for Java 17 even in `master`
branch for Apache Spark 3.3.0. Please stay tune this JIRA. We are doing our
best in the given timeline.
> Build and Run Spark on Java 17
> ------------------------------
>
> Key: SPARK-33772
> URL: https://issues.apache.org/jira/browse/SPARK-33772
> Project: Spark
> Issue Type: New Feature
> Components: Build
> Affects Versions: 3.3.0
> Reporter: Dongjoon Hyun
> Priority: Major
>
> Apache Spark supports Java 8 and Java 11 (LTS). The next Java LTS version is
> 17.
> ||Version||Release Date||
> |Java 17 (LTS)|September 2021|
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
