kylin-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Billy Liu (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (KYLIN-3125) Support SparkSql in Cube building step "Create Intermediate Flat Hive Table"
Date Thu, 25 Jan 2018 11:31:00 GMT

    [ https://issues.apache.org/jira/browse/KYLIN-3125?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16339104#comment-16339104
] 

Billy Liu edited comment on KYLIN-3125 at 1/25/18 11:30 AM:
------------------------------------------------------------

# Absolute path to beeline shell, can be set to spark beeline instead of the default hive beeline on PATH
 kylin.source.hive.beeline-shell=beeline

# given SparkSQL connects to the same Hive metastore.
 kylin.source.hive.enable-sparksql-for-table-ops=true
 kylin.source.hive.sparksql-beeline-shell=/path/to/spark-client/bin/beeline
 kylin.source.hive.sparksql-beeline-params=-n root- -hiveconf hive.security.authorization.sqlstd.confwhitelist.append='mapreduce.job.*|dfs.*' -u jdbc:hive2://localhost:10000
  


was (Author: yimingliu):
# Absolute path to beeline shell, can be set to spark beeline instead of the default hive beeline on PATH
kylin.source.hive.beeline-shell=beeline
 
# given SparkSQL connects to the same Hive metastore.
kylin.source.hive.enable-sparksql-for-table-ops=true
kylin.source.hive.sparksql-beeline-shell=/path/to/spark-client/bin/beeline
kylin.source.hive.sparksql-beeline-params=-n root --hiveconf hive.security.authorization.sqlstd.confwhitelist.append='mapreduce.job.*|dfs.*' -u jdbc:hive2://localhost:10000
 

> Support SparkSql in Cube building step "Create Intermediate Flat Hive Table"
> ----------------------------------------------------------------------------
>
>                 Key: KYLIN-3125
>                 URL: https://issues.apache.org/jira/browse/KYLIN-3125
>             Project: Kylin
>          Issue Type: New Feature
>            Reporter: liyang
>            Assignee: liyang
>            Priority: Major
>             Fix For: v2.3.0
>
>
> It is said that SparkSQL is much faster then Hive when used to create flat table.



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

Mime
View raw message