"Roughgarden has a suggestion that wouldn't be expensive to implement.
so do i. forget all this "normal physics" nonsense and just use hyperspace.
Before deciding which way to send information, he says, routers should consider not only which route seems the least congested, but also should take into account the effect that adding its own new messages will have on the route it has chosen.
i dunno, i don't think igrp would scale to the size of the internet. wasn't there a 1/(n^2) relationship between metadata size and network capacity as a function of total delay*bandwidth product in the whole system?
That would be, he says, 'just a bit altruistic' in that some routers would end up choosing routes that were not necessarily the fastest, but the average time for all users would decrease."
if somebody was looking for a problem to work on, i'd suggest that needing the shortest path to always have enough capacity makes planning crunchy. (which sounds like the same thing as quoted above, but really isn't.) -- Paul Vixie