when my websites seem slow I go to alertra.com and run a spot check. the
only thing is I don't know what's good and what's bad.

Are there guidelines about how fast a web server should respond? For
example, here is a spot check I just ran:

Time (US/Eastern) Checked From Result KBytes Secs Kbps
03/21/2006 21:58:51 Hong Kong CHINA OK 2.4 1.70 11.57
03/21/2006 21:58:50 Chicago USA OK 2.4 0.38 51.26
03/21/2006 21:58:50 Orlando USA OK 2.4 0.50 39.67
03/21/2006 21:58:50 Oklahoma City USA OK 2.4 0.43 46.19
03/21/2006 21:58:50 Las Vegas USA OK 2.4 0.49 40.24
03/21/2006 21:58:50 London UK OK 2.4 0.98 20.07
03/21/2006 21:58:51 Frankfurt GERMANY OK 2.4 1.27 15.53
03/21/2006 21:58:50 Atlanta USA OK 2.4 0.33 59.01


the range appears to be from 12 to 60 Kbps... is that good? It seems slow
from here in Vermont but I realize there are a variety of things that could
cause it to be slow on the user end, only one of which is the server itself.
That's why I use this alertra site but I don't know how to interpret it.

any advice?

thanks

In article <dOGdne4dELnsI73ZnZ2dnUVZ_tydnZ2d@adelphia.com>,
none@none.com says...
variables that could affect those results.

Addressing the range in Kbs, that is dependent on the number of requests
at that particular moment in time. Running that same test 1 second
earlier or 1 second later can give very different results.

It's also no surprise that the results are longer (or worse) when the
test is run from outside the US - more variables affecting the results.

OK, then... is there a good way to determine if your web server is slow?

In article <WqidnaSHBcqsUb3Z4p2dnA@adelphia.com>, none@none.com says...
would be considered fast, 30 - 50ms would be average. But as I said
before - many factors could skew that number.

There is no right answer here - slow is subjective. If you feel that
your server is serving pages too slowly at all times as opposed to other
sites on the internet during the same time period, then it is slow.

thanks... is there a website that can test response time objectively?

I think if my site is running slow and I complain to my host, they will give
excuses like it could be on my end. if there is a web site that measures
response time and I could send them the data, it would help.

thanks

In article <5qadnSS4pOHqTb3ZnZ2dnUVZ_tCdnZ2d@adelphia.com>,
none@none.com says...
www.analogx.com.

On Tue, 21 Mar 2006 23:22:42 -0500, gb put finger to keyboard and
typed:

What do you want to test? The thing that matters most to end users is
the load time, which is dependent on both the server and the content
of the page. Alexa gives a good comparison of load times as seen by
the end user, but it can't tell you why a site is slow (if it is
slow).

Essentially, there are five things that will affect the speed at which
a page loads:

* Server response time
* Outbound bandwidth from the server
* Inbound bandwidth to the user
* Amount of data in the page
* Browser render time

To get an objective test of any one of these, you would need to set up
a situation in which all the others were consistent and then alter the
variable you're trying to test. But that's very difficult, especially
since you don't have control over all of them.

If you think your site is slow, and it appears to be the server's
fault, then I'd suggest getting an account with one of the free
webspace providers and uploading a few copies of your pages there.
You've then got a comparison site to use when checking against your
main site - if the test pages on the free server load faster, then it
probably is the main server's fault. But even this can be affected by
how much back-end processing is taking place - scripting and database
calls will also affect performance, and you can't easily check for
these by using a duplicate page elsewhere.

Mark
--
Visit: http://www.MineOfUseless.info - everything you never needed to know!
Listen: http://www.goodge.co.uk/files/dweeb.mp3 - you'll love it!

right... in my situation, I hand code all my pages so the amount of data is
comparatively low (compared to pages created in WYSIWYG). And, I go to my
own websites very often so when they appear to be running slow, it's in
relation to how fast they usually run. That's why I think only the first
two in your list would be factors.

thanks

On Wed, 22 Mar 2006 10:00:43 -0500, gb put finger to keyboard and
typed:

OK. That's the sort of thing that a mirror site will allow you to
compare very easily.

Mark
--
Visit: http://www.CorporateContact.info - phone and email contacts for Amazon, Paypal, eBay and lots of other hard-to-contact organisations
Listen: http://www.goodge.co.uk/files/dweeb.mp3 - you'll love it!

On Tue, 21 Mar 2006 23:04:33 -0500, "gb" <none@none.com> posted
something that included:

If you have enough traffic, yet. Alexa.com (which is an Amazon site)
has scads of people using their toolbar. They spy on their users (with
the users' permission) and provide statistics on the various sites
their users visit.

Thus, if you have mucho traffic, you will have an average response
based on a lot of different users, some on broadband, some on dialup,
connecting from various places all over the world. It's not perfect,
of course; the Alexa toolbar probably gets used more by businessmen
more than housewives, more by adults than by kids, more by those who
speak English than those who speak other languages. On the other hand,
it's probably pretty good anyhow.

If you don't have enough traffic for Alexa to rate your site, you
aren't SOL. At various times, I've run wgets of the home page of
various sites from one server or another. Typically, I get results in
the 200-250 kbps range from full-service hosting companies, in the
100-120 kbps range for top sites running Windows, and speeds in the
10-30 range from sites like Geocities or from sites running off a
cable modem somewhere. On the other hand, the front page of Google
almost always is in 1-2 mbps range.

It's easy enough to do your own tests. Wget tells you not only the
time required to retrieve the file, but also the transfer rate.

Although the internet speeds up and slows down, I find that if site A
is faster than site B tonight at 7, it will also be faster tonight at
11, tomorrow morning at 3 and 10, and tomorrow afternoon at 1, 3, and
5 as well. Differences, yes, but not really big ones. But you'd be
foolish to take the word of a stranger, when it's easy enough to do
your own tests.







--
AmishHosting.com