[jira] [Commented] (JCR-3335) Reduce synchronized block scope when loading a node by id

["Julian Reschke (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/JCR-3335?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13292828#comment-13292828 ] 

Julian Reschke commented on JCR-3335:
-------------------------------------

(funny enough, I was looking at this as well :-)

Comments:

- we have that pattern in both getEntries(NodeImpl) and getEntries(NodeId), so we should fix them both

- once we have separated read from write operations, we may want to consider replacing the simple monitor by a ReadWriteLock...

Question: do we have test coverage for multithreaded access?
                
> Reduce synchronized block scope when loading a node by id
> ---------------------------------------------------------
>
>                 Key: JCR-3335
>                 URL: https://issues.apache.org/jira/browse/JCR-3335
>             Project: Jackrabbit Content Repository
>          Issue Type: Sub-task
>            Reporter: Alex Parvulescu
>         Attachments: CachingEntryCollector.java.patch
>
>
> This is a small(ish) change that takes the 'getNodeById' cll outside of the synchronized block that wraps the cache access.
> I see the load call as a heavy one, and the proposed change offers better concurrency behavior.
> In the tests that I ran, mostly related to queries I got a 20% improvement on average query response time.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira