You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hive.apache.org by "Zhihua Deng (Jira)" <ji...@apache.org> on 2025/07/07 01:07:00 UTC

[jira] [Updated] (HIVE-28800) Improve LongColumnStatsAggregator by preventing the overestimation of NDV

     [ https://issues.apache.org/jira/browse/HIVE-28800?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Zhihua Deng updated HIVE-28800:
-------------------------------
    Component/s: Standalone Metastore

> Improve LongColumnStatsAggregator by preventing the overestimation of NDV
> -------------------------------------------------------------------------
>
>                 Key: HIVE-28800
>                 URL: https://issues.apache.org/jira/browse/HIVE-28800
>             Project: Hive
>          Issue Type: Bug
>          Components: Standalone Metastore
>            Reporter: Seonggon Namgung
>            Assignee: Seonggon Namgung
>            Priority: Minor
>              Labels: pull-request-available
>             Fix For: 4.1.0
>
>
> When calculating the aggregated column statistics for a long-type column, the estimated NDV may sometimes exceed the total number of integers within the column’s value range. This estimated value is impossible and could lead to incorrect optimization decisions based on the statistics.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)