[arin-ppml] debunking the myth that Moore's law helps
On 12/17/2009 12:27 AM, Michel Py wrote:
>>> Michel's law 0.9 says: "A device that is twice as fast or has
>>> twice as much memory will consume twice as much bandwidth".
>> The doubling time may not be exactly 1.2 years (I've heard 18
>> months), but it definitely looks like a similar curve to me...
> Similar is the word. And although debating the actual value of the
> "multiplying factor" sounds interesting, I would welcome more feedback
> on the raw assertion that there is indeed a direct relationship between
> Moore's law and Internet bandwidth.
I've always looked at it from the supply side instead of the demand side
(although both probably contribute). The cost of transporting a bit is
a function of the cost to route it (electronics), and the cost to
transport it (optics). Routing costs seem to be tied to some version of
Moore's law. Although there are specialized components like TCAM that
don't observe the exact same factors as general purpose CPUs, they seem
to be on a similar exponential curve. Transport costs are somewhat more
volatile, especially with regard to the amount of dark fiber available
during and after the .com/telecom boom. But even given the now
relatively fixed amount of cheap dark fiber available, improvements in
link speeds and DWDM capacities have resulted in what appears to be an
exponential improvement in transport costs, which looks to continue at
least in the near and medium term.