] > Or, possible some small providers buy a multi-megabit circuit from a ] > large provider who gives them transit. The small provider then connects ] > at a single NAP and picks up bilateral peering sessions with a bunch ] > of people there. The result is offloading traffic from their ] > "transit link", which stands a good chance of being priced as a ] > "burstable" link. (pay for what you use) That gives the small ] > provider an economic incentive to operate in this manner.
] Quite a few CIX members operate this way. The interesting question in my ] mind is whether the "big guys" (defaultless nets, for the purposes of this ] discussion) think that this represents unfair competition or not.
We've a defaultless net, but I'm not sure that I'm considered a 'Big Guy'. Hell, we only route 1% of the internet, but maybe if I lost my aggregates I could be bigger ;)
The hidden metric that davec above doesn't consider is latency.
If I peer at a NAP, I forgo the latency my upstream 'multi-megabit circuit' incurs.
Hey, depends on your upstream provider and the NAP you're talking about... for some "Large" providers and at least one NAP, the reverse is true. =-) davec
participants (1)
-
Dave Curado