IP: Re: 100% per year, etc

[Dave Farber <farber () cis upenn edu>]

> Date: Sat, 2 Dec 2000 09:05:29 -0500 (EST)
> From: Andrew Odlyzko <amo () research att com>
> To: farber () cis upenn edu
> Cc: bill.st.arnaud () canarie ca, mo () UU NET
> Subject: Re: IP: 100% per year, etc
>
> Dave,
>
> Apropos Mike O'Dell's message that you sent out to the IP
> list on Thursday, he wrote:
>
>
> > I see people still don't really understand the difference
> > between offered load (measured as gigabits injected into
> > the edge of the network) and network capacity (measured in
> > gigabit-route-miles of trunking).
>
>
> This was apparently in reference to your note about my paper
> about Internet traffic growth myths,
>
>   <http://www.cisp.org/imp/november_2000/odlyzko/11_00odlyzko.htm>.
>
> However, that paper was not confused about the two measures he
> cites, and explicitly concentrated on the end-to-end traffic
> as measured in bytes (which is closely correlated with O'Dell's
> first measure, "offered load," and much more weakly related to
> network capacity).  The popular myths of Internet growth
> invariably speak of "traffic," as in the book "You Say You
> Want a Revolution" by the former FCC Chairman Reed Hundt, where
> he recently wrote, ``In 1999, data traffic was doubling every
> 90 days ...,'' or in two separate articles in the November 27
> issue of Fortune magazine.  Now I would claim that when people
> speak of car traffic, they do not refer to the number of lanes
> on a highway, or the length of a highway.  Similarly I would
> claim that when people talk of Internet traffic, their notion
> corresponds more closely to either byte volume or offered traffic,
> than to network capacity.  Thus by considering byte volume and
> showing that it is about doubling each year (which is consistent
> with Mike O'Dell's statement that on his network, offered load
> about doubles each year), I feel I am correcting a widely held
> misapprehension of what is happening on the Internet.
>
> Best regards,
> Andrew
>
>
> ************************************************************************
> Andrew Odlyzko                                      amo () research att com
> AT&T Labs - Research                                voice:  973-360-8410
> http://www.research.att.com/~amo                    fax:    973-360-8178
> ************************************************************************



For archives see: http://www.interesting-people.org/