[
https://issues.apache.org/jira/browse/DRILL-3549?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15197897#comment-15197897
]
Victoria Markman commented on DRILL-3549:
-----------------------------------------
I hope this setting will not be hard coded, but calculated based on cluster
settings + whatever else needs to be taken into consideration ...
> Default value for planner.memory.max_query_memory_per_node needs to be
> increased
> --------------------------------------------------------------------------------
>
> Key: DRILL-3549
> URL: https://issues.apache.org/jira/browse/DRILL-3549
> Project: Apache Drill
> Issue Type: Bug
> Components: Query Planning & Optimization
> Affects Versions: 1.2.0
> Reporter: Abhishek Girish
> Assignee: Deneche A. Hakim
> Priority: Critical
> Labels: usability
> Fix For: 1.7.0
>
>
> The current default value for planner.memory.max_query_memory_per_node is
> 2147483648 (2 GB). This value is not enough, given the addition of window
> function support. Most queries on reasonably sized data & cluster setup fail
> with OOM due to insufficient memory.
> The improve usability, the default needs to be increased to a reasonably
> sized value (could be determined based on Drill Max Direct Memory).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
