According to your reports I have written a small unit test, which helped me to identify the two problems, which are now fixed. However, as I

Thanks.

understand you, there are still problems, which my test doesn't expose.

I think this is where the confusion comes in.  Our application has a
load test driver and we found that ServletWebServer was a lot slower than
than just using WebServer.  I doubt that there is anything particular to
our application that identifies/causes this.

I was heading away from our application and trying to write a general
load test that just sent a few sample requests (simple one with just
a single field response, complex one with HashMap response, large one
with a big response and then add compression and SSL later) to see
the overall impact.  It could run as WebServer and then ServletWebServer
and potentially in a real servlet engine.

I don't know what ScalabilityTest is but it rings familiar to what I
was proposing.

Randall


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

Reply via email to