felix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Carsten Ziegeler (Resolved) (JIRA)" <j...@apache.org>
Subject [jira] [Resolved] (FELIX-3321) Improve implementation and reduce load on the service registry
Date Fri, 27 Jan 2012 12:37:39 GMT

     [ https://issues.apache.org/jira/browse/FELIX-3321?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Carsten Ziegeler resolved FELIX-3321.

    Resolution: Fixed

Committed a complete new implementation for handling the event handlers in revision 1236655

Instead of going to the service registry for each event, we now use a service tracker which
keeps track of registered event handlers. For each available handler, a proxy is created which
caches information about the handler like the topics, optional filter, timeout handling etc.
Once the first event is send to a handler the proxy gets the service from the registry and
keeps it until the service goes away or is modified.
This reduces the load of the service registry to a minimum

At the same time, the proxy allows us to optimize event handling and get rid of the special
filter caches for topics.

In addition I've improved the permission handling. If security is disabled, no permission
objects are created anymore

The whole thread handling has not changed at all - its the same as before

The new implementation passes the TCK (as before)
> Improve implementation and reduce load on the service registry
> --------------------------------------------------------------
>                 Key: FELIX-3321
>                 URL: https://issues.apache.org/jira/browse/FELIX-3321
>             Project: Felix
>          Issue Type: Improvement
>          Components: Event Admin
>    Affects Versions: eventadmin-1.2.14
>            Reporter: Carsten Ziegeler
>            Assignee: Carsten Ziegeler
>             Fix For: eventadmin-1.3.0
> The current implementation queries the service registry for each occuring event, gets
the services and then releases them. Especially with a high event volume this creates a very
high load on the service registry.

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