[jira] [Commented] (SPARK-12253) UnifiedMemoryManager race condition: storage can starve new tasks

["Apache Spark (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-12253?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15049918#comment-15049918 ] 

Apache Spark commented on SPARK-12253:
--------------------------------------

User 'andrewor14' has created a pull request for this issue:
https://github.com/apache/spark/pull/10240

> UnifiedMemoryManager race condition: storage can starve new tasks
> -----------------------------------------------------------------
>
>                 Key: SPARK-12253
>                 URL: https://issues.apache.org/jira/browse/SPARK-12253
>             Project: Spark
>          Issue Type: Bug
>          Components: Spark Core
>    Affects Versions: 1.6.0
>            Reporter: Andrew Or
>            Assignee: Andrew Or
>            Priority: Blocker
>
> The following race condition is possible with the existing code in unified memory management:
> (1) Existing tasks collectively occupy all execution memory
> (2) New task comes in and blocks while existing tasks spill
> (3) After tasks finish spilling, another task jumps in and puts in a large block, stealing the freed memory
> (4) New task still cannot acquire memory and goes back to sleep



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org