tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Lin <>
Subject benchmark
Date Fri, 18 Jan 2002 20:58:57 GMT

Greetings everyone, a quick thanks to the tomcat
developers for their hardwork.

I just spent two week performing benchmarks on tomcat
4.0.1 and I thought some people might like to know the
results I got.

My comparison was between Orion 1.5.2 and tomcat
4.0.1. I used JMeter to perform the test hitting an
application that gets XML data from another system
over the internet. The data was about 1.2K in size and
the parser used was xerces 2 and xalan.

By no means is my test comprehensive, but it does show
that tomcat 4.0.1 out performs orion 1.5.2 for 1-45
clients. The application is a real app which I can't
describe, but it performs two separate requests to an
external system to get XML data, then transforms it
into HTML. The system was a 900mhz p3 with 256m RAM,
jdk 1.3.1.

Tomcat vs Orion
1 client - 7% faster
5 client - 2% faster
15 clients - 9% faster
45 clients - 20% faster

The settings were the default setting for both. No
optimization or tweaking was done. With the default
settings for 90 and 135 clients, tomcat would fail.
Once I cranked up the min and max processors to 2-3x
the defaults, it was able to complete the tests.

I hope others find this interesting.


Do You Yahoo!?
Send FREE video emails in Yahoo! Mail!

To unsubscribe:   <>
For additional commands: <>
Troubles with the list: <>

View raw message