Sadly, I don't think it will be much use.
We have a heavy duty product that includes multiple (3 or 4) layers of
software running on a distributed architecture scattered across multiple OS
platforms. Part of that system used by the high level voice applications is
an API that is written in Java that sits on a Linux box and talks to an
Oracle server. We made a series of API calls from a test harness and
evaluated both internal timings in the API (the JSP/Java/Beans/Javascript
stuff) and external timings (at the client end, where the API results
eventually filter back across the network to. Our test harness drives this
whole setup to high levels of stress (enough to cause client side JVMs to
sometimes die on fast machines... never on the slow ones... due to resource
usage I suspect).
Though totally adequate for assessing the performance of our system, it will
be impossible for anyone else to use. Our results might be of interest (I'm
not sure if we'd consider those proprietary) but our method and technology
definitely would be proprietary (not to mention wwwwwwaaaayyyyy overkill for
most benchmarking).
Free Advice: Get JRE 1.3 from IBM or Sun (for its JVM/run-time environment)
- much faster than 1.2.2. The results were significantly improved timings.
Tomb
> -----Original Message-----
> From: G.Nagarajan [mailto:[EMAIL PROTECTED]]
> Sent: Monday, January 15, 2001 2:14 PM
> To: [EMAIL PROTECTED]
> Subject: RE: Tomcat Performance Questions "how was it tested?"
>
>
> Hi,
> can you tell how you tested the performance. I would like to apply the
> benchmark for my machine.
>
> Thanks
> Nagaraj.
>
> -----Original Message-----
> From: Stefan Langer [mailto:[EMAIL PROTECTED]]
> Sent: Monday, January 15, 2001 7:54 PM
> To: [EMAIL PROTECTED]
> Subject: RE: Tomcat Performance Questions
>
>
> One thing the documentation states is that you should use the
> AJP13 Protocol
> which is suppose to be faster. That connector is not configured in the
> default server.xml file. They use AJP12
> I have not tested this but you might want to give it a try.
>
> Stefan
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, email: [EMAIL PROTECTED]
>
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]
