[
https://issues.apache.org/jira/browse/KYLIN-5017?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17370998#comment-17370998
]
Congling Xia edited comment on KYLIN-5017 at 6/29/21, 2:32 AM:
---------------------------------------------------------------
Hi, [~vergilchiu]. AFAIK, cube level queue config is implemented by config
overwriting, too.
I mean you can overwrite configurations like
_kylin.engine.mr.config-override.mapreduce.job.queuename_ or
_kylin.engine.spark-conf.spark.yarn.queue_ at project level (on the page of
project management) to specify a queue for cubes in the same project.
was (Author: xiacongling):
Hi, [~vergilchiu]. AFAIK, cube level queue config is implemented by config
overwriting, too.
I mean you can overwrite configurations like
_kylin.engine.mr.config-override.mapreduce.job.queuename_ or
_kylin.engine.spark-conf.spark.yarn.queue_ ** ** at project level (on the page
of project management) to specify a queue for cubes in the same project.
> Support project level mapreduce queue config
> --------------------------------------------
>
> Key: KYLIN-5017
> URL: https://issues.apache.org/jira/browse/KYLIN-5017
> Project: Kylin
> Issue Type: Improvement
> Components: Job Engine
> Affects Versions: v3.1.2
> Reporter: vergilchiu
> Priority: Major
> Fix For: v2.0.0
>
>
> We have a lot of cubes in our compay , and for that we categorize cubes into
> diffent project based on user groups .
> We want this cubes to use diffent mapreduce queue . Although cube-level
> mapreduce queue config can meet our demands , we need to set up mapreduce
> queue for each cube. That's a lot of work.
> Now weed need a project-level mapreduce config , cube can use different
> queue for different project.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
