[
https://issues.apache.org/jira/browse/SPARK-21225?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16064682#comment-16064682
]
Apache Spark commented on SPARK-21225:
--------------------------------------
User 'JackYangzg' has created a pull request for this issue:
https://github.com/apache/spark/pull/18435
> decrease the Mem using for variable 'tasks' in function resourceOffers
> ----------------------------------------------------------------------
>
> Key: SPARK-21225
> URL: https://issues.apache.org/jira/browse/SPARK-21225
> Project: Spark
> Issue Type: Improvement
> Components: Spark Core
> Affects Versions: 2.1.0, 2.1.1
> Reporter: yangZhiguo
> Priority: Minor
> Original Estimate: 1h
> Remaining Estimate: 1h
>
> In the function 'resourceOffers', It declare a variable 'tasks' for
> storage the tasks which have allocated a executor. It declared like this:
> *{color:#d04437}val tasks = shuffledOffers.map(o => new
> ArrayBuffer[TaskDescription](o.cores)){color}*
> But, I think this code only conside a situation for that one task per core.
> If the user config the "spark.task.cpus" as 2 or 3, It really don't need so
> much space. I think It can motify as follow:
> {color:#14892c}*val tasks = shuffledOffers.map(o => new
> ArrayBuffer[TaskDescription](Math.ceil(o.cores*1.0/CPUS_PER_TASK).toInt))*{color}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
