For "normal" load, JMeter is a handy tool. I've seen tests with JMeter 
where (on the same machine) JMeter consumes more CPU/memory than the 
server-process.
For bigger load, it's better to use something like 
http://tsung.erlang-projects.org/.
With 100 users you have chosen a Tomcat "comfort"-zone test. Try the same 
test with 1000 users and give node the chance to use all cores and probably 
you'll get other numbers. BTW it wouldn't hurt to publish the full source 
in https://gist.github.com/ so we can play around with it.
If you have only 100 parallel connections at a time, stay with Tomcat.

-- 
Job Board: http://jobs.nodejs.org/
Posting guidelines: 
https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
You received this message because you are subscribed to the Google
Groups "nodejs" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/nodejs?hl=en?hl=en

Reply via email to