impala-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tim Armstrong (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (IMPALA-5084) Backend support for large rows in Sorter
Date Wed, 22 Mar 2017 00:44:41 GMT

    [ https://issues.apache.org/jira/browse/IMPALA-5084?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15935588#comment-15935588
] 

Tim Armstrong commented on IMPALA-5084:
---------------------------------------

As a temporary solution, it may be reasonable to just increase the default buffer size in
the sorter to match the maximum row size, given that the sorter's buffer requirements are
more modest than the joins and aggs, and there are typically fewer sorts in a plan.

> Backend support for large rows in Sorter
> ----------------------------------------
>
>                 Key: IMPALA-5084
>                 URL: https://issues.apache.org/jira/browse/IMPALA-5084
>             Project: IMPALA
>          Issue Type: Sub-task
>          Components: Backend
>    Affects Versions: Impala 2.6.0
>            Reporter: Tim Armstrong
>            Assignee: Tim Armstrong
>            Priority: Minor
>              Labels: resource-management
>
> See IMPALA-3208 for the context.
> Sorter::Run changes:
> * We can use a similar approach to that used for BufferedTupleStream as described in
IMPALA-5085
> Testing:
> Needs end-to-end tests exercising all operators with large operators



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Mime
View raw message