jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Schueler <g...@controltier.com>
Subject Webdav performance
Date Tue, 17 Feb 2009 18:41:47 GMT
Hi,
I work on an open-source project called ControlTier which consists of a few
webapps and a command-line client.  Both the client and one of the webapps
use a common Webdav server as a repository of files.  We use Jetty 6 as the
webapp container.

What can we do to maximize the performance of the most basic operations of
the Webdav component of Jackrabbit (PUT/GET)?

We don't need to use the more advanced features that the Jackrabbit content
repository has: search indexing, and other analysis of content.  We would
just like to use it as a simple file storage system with a Webdav
interface. We used to use the webdav server component builtin to Tomcat, but
we have moved to Jackrabbit. Jackrabbit has proved to be fairly robust and
stable, but for our purposes the performance has been a bit of a step
backwards.

I've looked through the documentation of the Jackrabbit configuration and
tried a few things to improve the performance: I removed the <SearchIndex>
portions of repository.xml and workspace.xml, and I removed the Zip/Xml
specific IO handlers in the servlet's config.xml. These changes don't seem
to make much improvement against some benchmark tests.

Are there other configuration options I can try to improve the performance?

Thanks,
Greg

-- 
Greg Schueler

ControlTier Software, Inc
greg@controltier.com
650-292-9660x709

http://www.controltier.com

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message