[
https://issues.apache.org/jira/browse/OAK-586?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13567512#comment-13567512
]
Marcel Reutegger commented on OAK-586:
--------------------------------------
Right, getNodes() does say the returning json tree must include the names of
the child nodes and the child count. On the positive side the names of the
child nodes can be limited by the client. E.g. if there are millions of child
nodes, a client can paginate through the child nodes if necessary. The child
count is IMO more problematic because it must be exact. This either means the
implementation needs to maintain the value somewhere and must update it
whenever a node is added or removed. This requires synchronization on this
value and causes contention. On the other hand an implementation, which choses
to not store the child node count is forced to count the potentially many child
nodes. I think it would be better to get rid of the child node count.
I'll follow up in OAK-231.
> Support for large number of child nodes
> ---------------------------------------
>
> Key: OAK-586
> URL: https://issues.apache.org/jira/browse/OAK-586
> Project: Jackrabbit Oak
> Issue Type: Improvement
> Components: mongomk
> Affects Versions: 0.6
> Reporter: Marcel Reutegger
>
> MongoMK currently stores nodes with a list of child node names. This does not
> scale when the number of child nodes increases.
--
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
