[
https://issues.apache.org/jira/browse/HIVE-17386?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sergey Shelukhin updated HIVE-17386:
------------------------------------
Attachment: HIVE-17386.patch
Adding tests, rebasing on top of the committed refactoring changes, and fixing
various scenarios.
While writing the tests, I realized the sync to avoid parallel requests for the
same session with different numbers is not so good. Would need to update that.
The non-"only" patch includes HIVE-17297
> support LLAP workload management in HS2 (low level only)
> --------------------------------------------------------
>
> Key: HIVE-17386
> URL: https://issues.apache.org/jira/browse/HIVE-17386
> Project: Hive
> Issue Type: Bug
> Reporter: Sergey Shelukhin
> Assignee: Sergey Shelukhin
> Attachments: HIVE-17386.only.patch, HIVE-17386.patch
>
>
> This makes use of HIVE-17297 and creates building blocks for workload
> management policies, etc.
> For now, there are no policies - a single yarn queue is designated for all
> LLAP query AMs, and the capacity is distributed equally.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
