[
https://issues.apache.org/jira/browse/OAK-4029?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15278063#comment-15278063
]
Julian Reschke commented on OAK-4029:
-------------------------------------
trunk: http://svn.apache.org/r1734941
1.4: http://svn.apache.org/r1743171
> RDBDocumentStore: optimize the bulk update method
> -------------------------------------------------
>
> Key: OAK-4029
> URL: https://issues.apache.org/jira/browse/OAK-4029
> Project: Jackrabbit Oak
> Issue Type: Improvement
> Components: rdbmk
> Reporter: Tomek Rękawek
> Assignee: Julian Reschke
> Fix For: 1.6, 1.5.0, 1.4.3
>
> Attachments: OAK-4029.patch
>
>
> It seems that the {{RDBDocumentStoreJDBC.update()}} method contains a
> redundant SELECT statement, between the UPDATE and INSERT phase. It's used to
> find out whether we should INSERT remaining documents or if they already
> exists in the DB. However, we may skip the middle SELECT and try to INSERT
> all documents with mod_count = 1.
> The second possible optimisation in this method is skipping the first bulk
> UPDATE operation if there are no documents to be updated.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
