Hi, Mladen. That's exactly what we expected, about 10% improvement.
We are runnning a cluster of 4 tomcats and we have a stress test batery , I'm not sure of how manu concurrent users it simulates, maybe that's the point. I've received new test data, when the tests are launched agaist only one tomcat (no cluster) it serves 4100 requests x per sencond, and when the four machines are running it only serves 4800!!!!.
I had SUSE9/x86_64, and it performs 10-20% faster then a regular connector. Of course, when you have a real web server and >100 concurrent clients with http/1.1 How did you figure out that the performance is twice lower the with the regular Tomcat? Regards, Mladen. --------------------------------------------------------------------- To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
-- Peace Sells.......... But Who's Buying? When all you have is a hammer, everything looks like a nail. --------------------------------------------------------------------- To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
