[jira] [Updated] (TEZ-3115) Shuffle string handling adds significant memory overhead

["Jonathan Eagles (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/TEZ-3115?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jonathan Eagles updated TEZ-3115:
---------------------------------
    Attachment: TEZ-3115.4.patch
                TEZ-3115.4-branch-0.7.patch

Thanks for the review, [~sseth]. Updated the patch for master and branch-0.7.

One thing to note is the difference between StringInterner.weakIntern vs String.intern is much smaller since jdk 7 when interned string were placed in the heap as opposed to perm gen.  Of course it is better to remain consistent in the project. Good catch.

http://java-performance.info/string-intern-in-java-6-7-8/

> Shuffle string handling adds significant memory overhead
> --------------------------------------------------------
>
>                 Key: TEZ-3115
>                 URL: https://issues.apache.org/jira/browse/TEZ-3115
>             Project: Apache Tez
>          Issue Type: Bug
>    Affects Versions: 0.7.0
>            Reporter: Jason Lowe
>            Assignee: Jonathan Eagles
>         Attachments: TEZ-3115.1.patch, TEZ-3115.2.patch, TEZ-3115.3-branch-0.7.patch, TEZ-3115.3.patch, TEZ-3115.4-branch-0.7.patch, TEZ-3115.4.patch
>
>
> While investigating the OOM heap dump from TEZ-3114 I noticed that the ShuffleManager and other shuffle-related objects were holding onto many strings that added up to over a hundred megabytes of memory.



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