Network latency is particularly resistant to Moore's law or anything
like it.
Well, sure; we're up to the point where speed-of-light is a substantial
fraction of the latency of long-haul links. I suspect - admittedly
without evidence - that if you subtract off the speed-of-light delay,
something rather like Moore's law will turn up.
When $DAYJOB opened an office in Thailand and people were complaining
about the network latency, I had to point out that something like half
the latency we were seeing was compelled by a combination of lightspeed
and going around (versus through) the planet. Our planet is something
like 140 light-milliseconds around, and we were seeing quarter-second
RTTs - and I don't for a moment think packets were taking great-circle
paths to get there.
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse(_at_)rodents-montreal(_dot_)org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
http://www.irtf.org/mailman/listinfo/asrg