[
https://issues.apache.org/jira/browse/HIVE-17651?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16189631#comment-16189631
]
Zoltan Haindrich commented on HIVE-17651:
-----------------------------------------
* it turned out that only in case of indexes the TS collects stats
* indexes are not supported for tez (I guess for spark also)
so..probably there is no issue arise from it.
> Profile TableScanOperator in case stats gathering is on
> -------------------------------------------------------
>
> Key: HIVE-17651
> URL: https://issues.apache.org/jira/browse/HIVE-17651
> Project: Hive
> Issue Type: Sub-task
> Reporter: Zoltan Haindrich
>
> It seems to me that there might be some overhead when the operator is
> gathering the stats;
> also check: it seems that in case vectorization is on ; the collected
> rowcount would be invalid
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
