You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@impala.apache.org by "Alexander Saydakov (Jira)" <ji...@apache.org> on 2021/12/13 23:25:00 UTC
[jira] [Resolved] (IMPALA-10956) datasketches UDFS: memory leak and merge overhead
[ https://issues.apache.org/jira/browse/IMPALA-10956?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Alexander Saydakov resolved IMPALA-10956.
-----------------------------------------
Resolution: Fixed
https://gerrit.cloudera.org/#/c/17869/
> datasketches UDFS: memory leak and merge overhead
> -------------------------------------------------
>
> Key: IMPALA-10956
> URL: https://issues.apache.org/jira/browse/IMPALA-10956
> Project: IMPALA
> Issue Type: Bug
> Components: Backend
> Affects Versions: Impala 4.0.0
> Reporter: Alexander Saydakov
> Priority: Minor
>
> I believe that there are memory leaks in aggregate UDFs that are using Apache Datasketches. Sketch or union objects are placed in a buffer to maintain the state of the aggregation, and later deallocated without calling destructors for those objects.
> Also there is unnecessary overhead during the merge phase of some of those UDFs, when an instance of a union is created for every merge operation and finalized by calling a quite expensive get_result() every time as well. This second issue is about performance, not a bug.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)