[jira] [Commented] (KUDU-1647) Improve range partition pruning and primary-key bound optimization with upper-bound range predicates on discrete typed columns

["Dan Burkert (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/KUDU-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15805991#comment-15805991 ] 

Dan Burkert commented on KUDU-1647:
-----------------------------------

[~honghaijie] implemented this for the C++ side in https://gerrit.cloudera.org/#/c/5360/.  Once those improvements have been ported over to the Java version we can close out this issue.

> Improve range partition pruning and primary-key bound optimization with upper-bound range predicates on discrete typed columns
> ------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: KUDU-1647
>                 URL: https://issues.apache.org/jira/browse/KUDU-1647
>             Project: Kudu
>          Issue Type: Sub-task
>          Components: client, tablet
>            Reporter: Dan Burkert
>
> We currently stop pushing predicates into the upper bound range partition key and pushing predicates into upper-bound primary key when a range predicate is encountered.  I _think_ (although we should probably think harder about this), that predicates could continue to be pushed if the exclusive range predicate value is decremented to make an inclusive upper bound predicate.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)