[
https://issues.apache.org/jira/browse/HIVE-12951?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15605750#comment-15605750
]
Xuefu Zhang commented on HIVE-12951:
------------------------------------
Besides more containers that are already active, the second run also benefits a
lot from JIT code generation which cannot be done by prewarning containers.
The right comparison is to measure the performance with and without container
prewarming to see if there is any benefit for your first run and make your
decision accordingly.
> Reduce Spark executor prewarm timeout to 5s
> -------------------------------------------
>
> Key: HIVE-12951
> URL: https://issues.apache.org/jira/browse/HIVE-12951
> Project: Hive
> Issue Type: Bug
> Components: Spark
> Affects Versions: 1.2.0
> Reporter: Xuefu Zhang
> Assignee: Xuefu Zhang
> Fix For: 2.1.0
>
> Attachments: HIVE-12951.patch, HIVE-12951.patch
>
>
> Currently it's set to 30s, which tends to be longer than needed. Reduce it to
> 5s, only considering jvm startup time. (Eventually, we may want to make this
> configurable.)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
