It might be a little simplistic, but we've always used a simple "100
user" metric: if we can simulate 100 simultaneous users on a site, or
webapp (in the sense of RIAs) and all the responses across the app come
back in reasonable times (which you can define variably as you see fit),
then we generally view that as good performance. Your 50-100ms response
time seems pretty reasonable... we usually tend to say something like
500ms AVERAGE across ALL requests aggregated from all virtual users ,
with none exceeding about a second. This is again pretty simplistic,
but it allows for a few things that take longer quite naturally to be
factored in and still ensure pretty reasonable performance across the board.
Frank
--
Frank W. Zammetti
Author of "Practical DWR 2 Projects"
and "Practical JavaScript, DOM Scripting and Ajax Projects"
and "Practical Ajax Projects With Java Technology"
for info: apress.com/book/search?searchterm=zammetti&act=search
Java Web Parts - javawebparts.sourceforge.net
Supplying the wheel, so you don't have to reinvent it!
My "look ma, I have a blog too!" blog: zammetti.com/blog
Leon Rosenberg wrote:
Hi all,
recently I was asked by someone to define performance requirements for
a site without even knowing it (the site).
The task was to deliver standard requirements valid for most sites (or
portal-like sites).
I answered that it's probably impossible without knowing the type of
media delivered, whether its web 2.0 or 1.0, and so on...
but I think as long as the webserver starts delivering content in
between 50 and 100 ms (measured at webserver) the site should be ok.
What do you think could be the site-independent performance criterion?
best regards
Leon
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
