Re: types of traffic in tcp congest control

Joe Touch <touch@ISI.EDU> Wed, 15 May 2002 05:48 UTC

Message-ID: <3CE1F6BE.6080908@isi.edu>
Date: Tue, 14 May 2002 22:48:46 -0700
From: Joe Touch <touch@ISI.EDU>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.0.0+) Gecko/20020509
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Anumita Biswas <BAnumita@novell.com>
Cc: alan@lxorguk.ukuu.org.uk, jinw@comp.leeds.ac.uk, tcp-impl@lerc.nasa.gov, dotis@sanlight.net
Subject: Re: types of traffic in tcp congest control
References: <sce1998e.076@prv-mail25.provo.novell.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-tcp-impl@grc.nasa.gov
Precedence: bulk
Status: RO
Content-Length: 633
Lines: 18


Anumita Biswas wrote:
> Is it not true that overprovisioning is never a long term solution, as
> bandwidth is never ever enough? As Alan says, "Usage expands to fit
> network bandwidth".

If you have a DSL line, track its usage. You will find that ~1Mbps is 
more than sufficient most of the time.

Yes, applications sometimes catch up. However, whenever a resource is 
scarce the assumption is that usage will always catch up. The arguments 
held in the past for disk drives, memory capacity, and CPU speed. Most 
are, for most purposes, overprovisioned very easily for all but a small 
fraction of the time for most users.

Joe