Re: [>Htech] Cornell News: How small worlds work

Date view Thread view Subject view Author view

From: Dave Long (
Date: Thu Sep 14 2000 - 12:05:33 PDT

> Exactly which distant nodes to discover/remember would be determined by a
> broadcast probe and consistent hashing: that is, bias the probe message to
> go "further" (more geographic hops) as it finds nodes "closer" (on the
> hashed unit circle).

I'm not sure how you are arranging that distance on the
hashed unit (circumference?) circle has anything to do
with hopcount, as the hashing ought to remove residual
information about geography.

After one has discovered a certain number of nodes, maybe
taking "near" hashed ones, and checking their hopcounts
would give some idea of the radius of the network? Could
it be useful to determine when one has found "distant"
enough nodes?

I'd think instead that one would run the broadcast probe,
and by somehow exchanging information about neighborhoods
determine which nodes to keep in the routing set. After
all, the goal is for each node to have a large number of
nodes within its clique, but also a significant number of
nodes that have substantially differing cliques.


Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Sep 14 2000 - 11:55:26 PDT