[10358] in North American Network Operators' Group
Re: Keynote/Boardwatch Internet Backbone Index A better test!!!
daemon@ATHENA.MIT.EDU (Michael Dillon)
Mon Jun 30 18:00:36 1997
In-Reply-To: <Pine.LNX.3.91.970627205910.16719F-100000@zen.cypher.net>
Date: Mon, 30 Jun 1997 14:53:44 -0700
To: nanog@merit.edu
From: Michael Dillon <michael@priori.net>
At 9:00 PM -0400 6/27/97, Ben Black wrote:
>i see, so i am to understand you wish us to simply *trust* that your
>results are valid based on vague references to your methodology?
>
>hey everyone, i have just completed a study that indicates in some
>circumstances a 28.8 modem is faster than a DS-3! no details at this
>time, but i swear it is all true!
This isn't as silly as you think.
At the February NANOG, Van Jacobsen showed the results of some research in
which they discovered some situations where web pages could be retrieved
over a 28.8k modem link faster than over a workstation connected to
Ethernet, connected to a T1 line.
Just remember, that TCP flows are not water flows and aggregated TCP flows
in a big pipe do not behave at all like aggregated water flows in a big
pipe.
********************************************************
Michael Dillon voice: +1-415-482-2840
Senior Systems Architect fax: +1-415-482-2844
PRIORI NETWORKS, INC. http://www.priori.net
"The People You Know. The People You Trust."
********************************************************
