You are viewing a plain text version of this content. The canonical link for it is here.

Posted to oak-issues@jackrabbit.apache.org by "Marcel Reutegger (JIRA)" <ji...@apache.org> on 2013/09/24 16:54:02 UTC

[jira] [Commented] (OAK-1034) OOME running Oak benchmark suite

    [ https://issues.apache.org/jira/browse/OAK-1034?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13776362#comment-13776362 ] 

Marcel Reutegger commented on OAK-1034:
---------------------------------------

bq. CreateManyChildNodesTest on MongoMK

The issue isn't just about memory usage, but also MongoDB document size. To recap, the test creates 10'000 child nodes and calls save. This results in 40 commits to a branch, followed by a merge. Every commit contains a list of child node names because the node type is nt:unstructured and the :childOrder property is maintained with every commit. This results in very big MongoDB documents even though we now have document splitting in place. The current implementation only takes the number of commits into account to trigger a split (1000 commits). This needs to be improved to split earlier as soon as the document reaches a given size threshold.
                
> OOME running Oak benchmark suite
> --------------------------------
>
>                 Key: OAK-1034
>                 URL: https://issues.apache.org/jira/browse/OAK-1034
>             Project: Jackrabbit Oak
>          Issue Type: Bug
>          Components: mongomk, run
>            Reporter: Michael Dürig
>            Assignee: Marcel Reutegger
>
> Several of the benchmarks result in an OOME running with {{-Xmx512M}}

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira