[ 
https://issues.apache.org/jira/browse/KUDU-2645?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Mike Percy updated KUDU-2645:
-----------------------------
    Code Review: https://gerrit.cloudera.org/c/12205/

> Diff scanner should perform a merge on the rowset iterators at scan time
> ------------------------------------------------------------------------
>
>                 Key: KUDU-2645
>                 URL: https://issues.apache.org/jira/browse/KUDU-2645
>             Project: Kudu
>          Issue Type: New Feature
>          Components: tablet
>            Reporter: Mike Percy
>            Assignee: Mike Percy
>            Priority: Major
>
> In order to perform a diff scan we will need the MergeIterator to ensure that 
> duplicate ghost rows are not returned in cases where a row was deleted and 
> flushed, then reinserted into a new rowset during the time period covered by 
> the diff scan. In such a case, only one representation of the row should be 
> returned, which is the reinserted one.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to