You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Avi minsky (JIRA)" <ji...@apache.org> on 2018/11/21 19:41:00 UTC
[jira] [Updated] (SPARK-26143) Shuffle shuffle default persist type
[ https://issues.apache.org/jira/browse/SPARK-26143?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Avi minsky updated SPARK-26143:
-------------------------------
Description: Currently developer can set storage level explicitly only on persist command but shuffling can occur in many cases (group by, join, etc..) why can we set a default persist type, this can be helpful in many cases (also will automatically allow replication of shuffle blocks if set to memory_only_2, for example) (was: Currently developer can set persist type explicitly only on persist command but shuffling can occur in many cases (group by, join, etc..) why can we set a default shuffle mode, this can be helpful in many cases (also will automatically allow replication of shuffle blocks if set to memory_only_2, for example))
> Shuffle shuffle default persist type
> ------------------------------------
>
> Key: SPARK-26143
> URL: https://issues.apache.org/jira/browse/SPARK-26143
> Project: Spark
> Issue Type: New Feature
> Components: Shuffle
> Affects Versions: 2.3.0
> Reporter: Avi minsky
> Priority: Major
>
> Currently developer can set storage level explicitly only on persist command but shuffling can occur in many cases (group by, join, etc..) why can we set a default persist type, this can be helpful in many cases (also will automatically allow replication of shuffle blocks if set to memory_only_2, for example)
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org