jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jukka Zitting (Commented) (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (JCR-3117) Stats for the PersistenceManager
Date Thu, 20 Oct 2011 09:36:10 GMT

    [ https://issues.apache.org/jira/browse/JCR-3117?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13131482#comment-13131482

Jukka Zitting commented on JCR-3117:

> the 2 counts are just incremental operations on AtomicLong, I don't really see the added
value in using TS;

There's two main benefits: 1) we get a full time series of the collected values, and 2) we
don't need the extra interface or custom code for accessing this piece of information.

> the 'cache miss avg duration' : AFAIK the TS impl does not do averages currently, if
it will, I'm open for using it.

The average duration is easily calculated by dividing total duration by count per second.
It's not currently supported by the TS implementation, but will be easy enough to add.
> Stats for the PersistenceManager
> --------------------------------
>                 Key: JCR-3117
>                 URL: https://issues.apache.org/jira/browse/JCR-3117
>             Project: Jackrabbit Content Repository
>          Issue Type: Sub-task
>          Components: jackrabbit-core
>            Reporter: Alex Parvulescu
>            Assignee: Alex Parvulescu
>            Priority: Minor
>             Fix For: 2.3.2
>         Attachments: 0001-JCR-3117-Stats-for-the-PersistenceManager.patch, JCR-3117.patch
> Statistics for the PersistenceManager impl that cover: 
>  - bundle cache access count, 
>  - bundle cache miss count, 
>  - bundle cache miss avg duration (this avg includes the penalty of having to load from
the underlying storage / can be interpreted as avg read latency as there is no cache involved)

>  - bundle writes per second
> What it doesn't cover is :
>  - number of bundles
>  - size of workspace
> as these are values that are expensive to compute on demand, and caching them would imply
being able to store the values (which is not possible currently).

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


View raw message