stephen at clark.net
Thu Aug 29 21:16:38 UTC 1996
On Thu, 29 Aug 1996, Perry E. Metzger wrote:
> Vadim Antonov writes:
> > Perry E. Metzger <perry at piermont.com> wrote:
> > >Raw bandwidth is inherently cheap. Undersea lines are even fairly
> > >cheap, except nasty government price fixing keeps them
> > >expensive. Switching equipment is expensive right now but Moore's Law
> > >takes care of that over the years.
> > Excuse me, but bandwidth demand doubles in about half year,
> > while Moore's law is that semiconductor capacity doubles every
> > 2 years. There's no indication that this will change any time
> > soon.
> > Nice brick wall :)
> I doubt it. The curve is racing upward now because of all the people
> who are suddenly connecting to the net. Once a large fraction of them
> are on the curve will slow dramatically -- demand for bandwidth will
> continue to increase, but only as fast as the customers can eat it,
> which is by definition related to how fast their equipment runs.
> This whole thing is very reminiscent of what happened with fax machine
> production a decade or so ago, and what happened with phone installs
> about a century ago.
> There will be some problems between now and the time things slow down,
Networks are by definition designed for one computer to talk with many
other computers. Under the ideal situation the computers would talk to
each other at the same speed it talks to its self (bus bandwidth, memory
speed) creating a single virtual computer. Due to the engineering
constraints of doing this, using the bandwidth that is available,
applications have been written that are "network specific" eg. slow. This
constraint will always drive bandwidth to higher speeds. Thus it follows
Moore's Law as chips get faster, so will computers and conversly -
bandwidth. There is no end in sight for the need of more bandwidth.
Stephen Balbach "Driving the Internet To Work"
VP, ClarkNet due to the high volume of mail I receive please quote
info at clark.net the full original message in your reply.
More information about the NANOG