Speedtest site accuracy [was: Bandwidth issues in the Sprint network]

Tue Apr 8 19:20:26 UTC 2008

We have an test server inside our network that we have customers test again.
We tell customers that we can only control our network -- beyond our
upstream routers it's best-effort only.

That said, if there is a real performance issue upstream we do our best to
assist or point the customer in the right direction.

Frank

-----Original Message-----
From: owner-nanog at merit.edu [mailto:owner-nanog at merit.edu] On Behalf Of
Scott Weeks
Sent: Tuesday, April 08, 2008 1:16 PM
To: nanog at merit.edu
Subject: Speedtest site accuracy [was: Bandwidth issues in the Sprint
network]

--- mevans at alphatheory.com wrote:

Try using the Java test on DSLReports rather than the Flash based test. I've
found it to be much more accurate. I also receive the message about
compression being used when I test with the flash test. I think it may be a
bug.
---------------------------------------

This brings up a PITA point for me.  Recently, I have seen a rash of
"Speedsite test server at <location> says blah, blah, blah" tickets finally
reach me and I am telling everyone they're not an accurate way to measure
network performance.  I notice that at least some are just sending text in
Latin.

To other medium-sized eyeball network providers (I'm defining medium size as
50-150K DSL/Cable connections and 50-1500 leased line customers): are you
seeing this and what do you tell your customers?

scott