[
https://issues.apache.org/jira/browse/SPARK-14836?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15253254#comment-15253254
]
Apache Spark commented on SPARK-14836:
--------------------------------------
User 'jerryshao' has created a pull request for this issue:
https://github.com/apache/spark/pull/12597
> Zip local jars before uploading to distributed cache
> ----------------------------------------------------
>
> Key: SPARK-14836
> URL: https://issues.apache.org/jira/browse/SPARK-14836
> Project: Spark
> Issue Type: Improvement
> Components: YARN
> Affects Versions: 2.0.0
> Reporter: Saisai Shao
> Priority: Minor
>
> Currently if neither {{spark.yarn.jars}} nor {{spark.yarn.archive}} is set
> (by default), Spark on yarn code will upload all the jars in the folder
> separately into distributed cache, this is quite time consuming, and very
> verbose, instead of upload jars separately into distributed cache, here
> changes to zip all the jars first, and then put into distributed cache.
> This will significantly improve the speed of starting time, in my local
> machine, it could save around 5 seconds for the starting period, not to say a
> real cluster.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
