Mailing-List: contact tomcat-user-help@jakarta.apache.org; run by ezmlm
Message-ID: <39774CAC.44C18203@eng.sun.com>
Date: Thu, 20 Jul 2000 12:02:05 -0700
From: "Craig R. McClanahan" <Craig.McClanahan@eng.sun.com>
MIME-Version: 1.0
To: tomcat-user@jakarta.apache.org
Subject: Re: Tomcat benchmarks from orionserver.com, Please comment. Important
References: <511F5E23FD70D111BA840060083424E5B6DBFB@pt73.peacetech.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

"Roytman, Alex" wrote:

> Craig,
>
> Sorry if my question sounded silly. Of course "fair" was meant as "whether
> this test gives good representation of performance difference in
> implementations tested" given that performance test can be tricky task.
> I have *no doubts* that people who ran the test are honest a ethical /is it
> "fair" to assume that I was trying to insult them? :-( /
>

I didn't think you were ... but I've seen people ask this kind of a question
from that kind of motive before.

> Actually, my apps do not suffer much from Tomcat's performance due to the
> fact that tomcat's overhead is small to compare with "heavy stuff like XSLT"
> I run inside of my Servlets. And I am using version 3.2 by the way.
>

Thus reinforcing the point I think is important here -- benchmarks are only
useful as a behavior predictor if they adequately mirror the behavior of the
application you will actually run.  And "adequately" is definitely in the eyes
of the beholder.

>
> Thank you
>
> Alex
>

Craig McClanahan