[51845] in North American Network Operators' Group
Re: Baltimore train tunnels (was Re: Vulnerbilities of Interconnection)
daemon@ATHENA.MIT.EDU (William B. Norton)
Sun Sep 8 19:34:21 2002
Date: Sun, 08 Sep 2002 16:21:06 -0700
From: "William B. Norton" <wbnorton@pacbell.net>
To: Sean Donelan <sean@donelan.com>, nanog@merit.edu
Errors-To: owner-nanog-outgoing@merit.edu
At 09:47 PM 9/7/2002 -0400, Sean Donelan wrote:
>Unlike phone calls, TCP traffic doesn't occur in fixed bandwidth
>increments. TCP traffic, 90% of Internet traffic, is elastic. By design,
>TCP adjusts the traffic rate to keep the bottleneck congested. As the
>bottleneck moves, traffic reacts by increasing or decreasing the rate to
>match the available capacity. This feedback occurs independently of what
>is happening on nearby traffic paths. Even if there is available
>capacity on elsewhere, the current Internet design is not very good at
>using it. Some people view this as an inefficient use of available
>capacity, other people view it as a self-protective mechanism.
Thank Goodness for well-behaved applications, right? ( Misbehaving TCP
stacks and UDP-based apps don't obey these back off rules. ) I remember Van
Jacobson gave a presentation back in 1997 that spoke about the problems
with applications that didn't exhibit these characteristics:
http://www.academ.com/nanog/october1997/
It would be interesting to see some recent verification that well-behaved
TCP-apps are the norm on the Internet...any data out there in this regard?
Bill
