Re: types of traffic in tcp congest control

Alan Cox <alan@lxorguk.ukuu.org.uk> Wed, 15 May 2002 11:42 UTC

Subject: Re: types of traffic in tcp congest control
To: touch@ISI.EDU (Joe Touch)
Date: Wed, 15 May 2002 12:42:51 +0100 (BST)
Cc: BAnumita@novell.com (Anumita Biswas), alan@lxorguk.ukuu.org.uk, jinw@comp.leeds.ac.uk, tcp-impl@lerc.nasa.gov, dotis@sanlight.net
In-Reply-To: <3CE1F6BE.6080908@isi.edu> from "Joe Touch" at May 14, 2002 10:48:46 PM
X-Mailer: ELM [version 2.5 PL6]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-Id: <E177xBD-0001iq-00@the-village.bc.nu>
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Sender: owner-tcp-impl@grc.nasa.gov
Precedence: bulk
Status: RO
Content-Length: 503
Lines: 12

> 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.

Only because the network core is too slow and the charging models for services
are underdeveloped. Porn on demand will fill plenty of 1Mbit DSL pipes once
the infrastructure can handle it.