30 seconds? No way is anyone going to wait 30 seconds fora page to download
these days, unless it's something they are really interested in seeing.
Still, 30 seconds would even be pushing it then. Aim for 10 or less.
Also, the problem with a few 100ms is, what happens when you multiply that
by 1,000 people? It adds up fast. For example: I used to have a page that
took 2.5 seconds to generate. That doesn't sound too bad. 2.5 seconds
falls well within your 30 second guideline, but what happens if just 10
people hit this page at once? All of a sudden, you are looking at 25
seconds of processing time just for 10 users. If the traffic is steady
enough, your machine will get further and further behind until it starts
timing out, or dies. Of course I managed to knock that time down with Views
and Stored Procedures and what-not, but still .. it's a good example.
As for your traffic example, I have a better one for you. Say you are
driving down the freeway and way up the road there is a fender bender.
Well, of course, everyone who drives by slows down just a tiny bit to have a
look Well, that little slow down has a ripple effect that goes waaaaay back
down the line of traffic and each little slow down adds to the ripple.
Eventually, what always happens is a traffic jam about 3 miles back. Living
around Charlotte, I see this almost every day.
Same thing with your CF Server.
Todd Ashworth
----- Original Message -----
From: "Patrick McElhaney" <[EMAIL PROTECTED]>
To: "Fusebox" <[EMAIL PROTECTED]>
Sent: Thursday, January 25, 2001 10:17 AM
Subject: RE: Page execution time too large?
| When you travel across town, does it really matter whether
| a particular traffic light is 10 seconds or 60 seconds?
|
| I don't think the page execution time is worth worrying
| about, because when you look at the total time from the
| user's perspective (request + execution time + response
| + page rendering), the execution time accounts for only a
| very small percentage of the round trip. And the user isn't
| going to notice the difference of a few 100ms.
|
| What I believe is far more important is making sure that
| your application can handle the load, by locking access
| to shared-scope variables[1], caching queries,
| understanding transactions, etc.
|
| Also, keep in mind, Cold Fusion is optimized for development
| speed, not execution speed. It's not worth the trouble
| to try to optimize the code. If you have a very processor-
| intensive algorithm, rather write it in a language designed
| for speed, like C++. CF provides many ways to tap into other
| languages (stored procedures, COM/CORBA, CFX, servlets, etc)
| for this purpose.
|
| Note that hardware is getting cheaper every year, while
| programmers are getting more expensive. :)
|
| In the days when the web was mostly static, I believe a
| common guideline was that a page should take no more than
| 30 seconds on a 28.8k modem to load (with images and
| everything). Nowadays, the lowest common denominator is
| more like 56k, so if you're looking for a guideline, make
| sure your pages can load on a 56k modem in 30 seconds,
| you're doing well.
|
| Patrick
|
| [1] http://www.allaire.com/handlers/index.cfm?id=17318&method=full
|
|
| > -----Original Message-----
| > From: Kola Oyedeji [mailto:[EMAIL PROTECTED]]
| > Sent: Wednesday, January 24, 2001 10:29 AM
| > To: Fusebox
| > Subject: Page execution time too large?
| >
| >
| > I wondered if there are any guidelines for a maximum amount of
| > miliseconds a
| > page should take to process. I have a page which takes 150milliseconds
to
| > execute is this acceptable?
| >
| > What do others recommend should be the maximum execution time of a
page??
| >
| >
|
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Structure your ColdFusion code with Fusebox. Get the official book at
http://www.fusionauthority.com/bkinfo.cfm
Archives: http://www.mail-archive.com/[email protected]/
Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
