I realized that the numbers that I posted before weren't accurate and I wanted to amend them.
Our 75 user test generates 745 transactions/minute and our 150 user test generates 1500 transactions/minute (not the 5000 and 10000 numbers listed before). The previous version of our application, running on Weblogic Server, was able to handle loads up to 2200 transactions/minute on the same hardware. We also tested to see if the problem was related to the maxkeepalive parameter. We tested with this feature disabled (maxKeepAlive=1), but unfortunately this didn't seem to have any affect. So anyone have any thoughts on this one? -- -Rob Heim
