hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sahil Takiar (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-19937) Intern JobConf objects in Spark tasks
Date Fri, 29 Jun 2018 01:29:00 GMT

    [ https://issues.apache.org/jira/browse/HIVE-19937?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16527015#comment-16527015
] 

Sahil Takiar commented on HIVE-19937:
-------------------------------------

Thanks for the review Xuefu!

[~misha@cloudera.com] could you take a quick look too, just want to make sure I am doing this
correctly. The high level explanation is that in Hive on Spark, a single Spark executor (a
single JVM) can run multiple Spark tasks in parallel. Each task creates its own copy of {{JobConf}}
(basically all the Hive properties). This can cause significant overhead in the JVM, so this
patch basically interns all the entries in {{JobConf}}.

> Intern JobConf objects in Spark tasks
> -------------------------------------
>
>                 Key: HIVE-19937
>                 URL: https://issues.apache.org/jira/browse/HIVE-19937
>             Project: Hive
>          Issue Type: Improvement
>          Components: Spark
>            Reporter: Sahil Takiar
>            Assignee: Sahil Takiar
>            Priority: Major
>         Attachments: HIVE-19937.1.patch
>
>
> When fixing HIVE-16395, we decided that each new Spark task should clone the {{JobConf}}
object to prevent any {{ConcurrentModificationException}} from being thrown. However, setting
this variable comes at a cost of storing a duplicate {{JobConf}} object for each Spark task.
These objects can take up a significant amount of memory, we should intern them so that Spark
tasks running in the same JVM don't store duplicate copies.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message