You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@impala.apache.org by "Tim Armstrong (JIRA)" <ji...@apache.org> on 2017/08/24 21:54:00 UTC
[jira] [Resolved] (IMPALA-4679) Impala to set query mem_limit
automatically for compute stats
[ https://issues.apache.org/jira/browse/IMPALA-4679?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Tim Armstrong resolved IMPALA-4679.
-----------------------------------
Resolution: Duplicate
I don't think we want to do a one-off for compute stats - the general solution we're taking is to impose better memory constraints on all operators. Then we'll be able to have better admission control policies.
> Impala to set query mem_limit automatically for compute stats
> -------------------------------------------------------------
>
> Key: IMPALA-4679
> URL: https://issues.apache.org/jira/browse/IMPALA-4679
> Project: IMPALA
> Issue Type: Improvement
> Components: Backend, Frontend
> Affects Versions: Impala 2.8.0
> Reporter: Alan Choi
> Assignee: Marcel Kornacker
>
> With Multi-threading turned on for Compute Stats, the memory usage in the SCAN is a lot more deterministic. In fact, the memory usage in SCAN and AGG can be pre-determined in the planner. Therefore, it is feasible to set the mem_limit automatically for compute stats query.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)