[
https://issues.apache.org/jira/browse/HIVE-28646?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17902672#comment-17902672
]
Qiheng He commented on HIVE-28646:
----------------------------------
[~lisoda] To be honest, I’m still not sure how to change the documentation,
because Hive 4.1.0 has not been released yet, and it seems that there is
nothing I need to do.
> Multiple documents for `hive.execution.engine` appear to be out of date
> -----------------------------------------------------------------------
>
> Key: HIVE-28646
> URL: https://issues.apache.org/jira/browse/HIVE-28646
> Project: Hive
> Issue Type: Improvement
> Security Level: Public(Viewable by anyone)
> Reporter: Qiheng He
> Priority: Major
>
> - According to https://github.com/apache/hive/pull/4973 , the default value
> of *hive.execution.engine* should now be {*}tez{*}.
> - But
> https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=27842758#ConfigurationProperties-hive.execution.engine
> still records that the default value of *hive.execution.engine* is *mr*.
> {code:none}
> hive.execution.engine
> Default Value: mr (deprecated in Hive 2.0.0 – see below)
> Added In: Hive 0.13.0 with HIVE-6103 and HIVE-6098
> Chooses execution engine. Options are: mr (Map Reduce, default), tez (Tez
> execution, for Hadoop 2 only), or spark (Spark execution, for Hive 1.1.0
> onward).
> While mr remains the default engine for historical reasons, it is itself a
> historical engine and is deprecated in the Hive 2 line (HIVE-12300). It may
> be removed without further warning.
> See Hive on Tez and Hive on Spark for more information, and see the Tez
> section and the Spark section below for their configuration properties.
> {code}
> - At the same time, the *Hive on Spark* documentation also records that the
> default value of *hive.execution.engine* is *mr*. This seems too
> disconnected. See
> https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=42567714#HiveonSpark-2.1ANewExecutionEngine
> .
> {code:none}
> We will introduce a new execution, Spark, in addition to existing MapReduce
> and Tez. To use Spark as an execution engine in Hive, set the following:
> set hive.execution.engine=spark;
> The default value for this configuration is still “mr”.
> {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
