You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@hive.apache.org by "Xuefu Zhang (JIRA)" <ji...@apache.org> on 2015/06/01 23:17:35 UTC

[jira] [Updated] (HIVE-10671) yarn-cluster mode offers a degraded performance from yarn-client [Spark Branch]

     [ https://issues.apache.org/jira/browse/HIVE-10671?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Xuefu Zhang updated HIVE-10671:
-------------------------------
    Fix Version/s:     (was: spark-branch)
                   1.3.0

> yarn-cluster mode offers a degraded performance from yarn-client [Spark Branch]
> -------------------------------------------------------------------------------
>
>                 Key: HIVE-10671
>                 URL: https://issues.apache.org/jira/browse/HIVE-10671
>             Project: Hive
>          Issue Type: Bug
>          Components: Spark
>            Reporter: Xuefu Zhang
>            Assignee: Rui Li
>             Fix For: 1.3.0
>
>         Attachments: HIVE-10671.1-spark.patch, HIVE-10671.2-spark.patch
>
>
> With Hive on Spark, users noticed that in certain cases spark.master=yarn-client offers 2x or 3x better performance than if spark.master=yarn-cluster. However, yarn-cluster is what we recommend and support. Thus, we should investigate and fix the problem. One of the such queries is TPC-H  22.



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