Filip Hanik wrote:
How heavy of a load does this generate, in terms of # of req/s?

execute: java -cp testclient.jar;commons-httpclient.jar;commons-logging.jar org.apache.catalina.cluster.test.client.MemTestClient http ://192.168.0.103:7080/index.jsp 100 100000 1000 false

java -cp testclient.jar;commons-httpclient.jar;commons-logging.jar
org.apache.catalina.cluster.test.client.MemTestClient get-url nrofthreads
nrofrequests sleep-between-request usehttp11

Maybe this isn't useless after all: I've finally found a benchmark where 5.0.17 wipes the floor with 4.1.29 (I did increase the maxProcessors to 150; the bench with those settings obviously won't work with only 75 processors).


For a static request (2 KB; /tomcat.gif) the average response time is 4 ms (compared to 10 ms), less memory (about 20 MB less; of course, this is more heap due to more GC, not actually used memory), and less CPU time (0-3% compared to 12-14%). Somehow, ab doesn't show that kind of difference (too bad, I would have been happy).

Rémy



---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to