Re: Benchmarking Dynamic Web XSLT

On a P2 350 Mhz, I'm able to do 10 gets/second. I use Apache JServ, XT and a
database.  The XSLT translation is not the problem.  The overhead appears to
be in the communication between apache, the app server and the database
engine.

10 tps is 100ms per request with only about 12ms it in the XSLT engine (and
it's written in Java). Cached, compiled XT is very fast.

These systems are very sensitive to minor changes. If I mess up anything I
drop to 3 tps. I'd like to be at 20 tps on this hardware.

Jon Smirl
jonsmirl@xxxxxxxxxxxx



 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list