[e2e] Overly Overlay; Peer to peer is commonplace
David P. Reed
dpreed at reed.com
Mon Jan 7 17:14:26 PST 2002
Considering wired networks, it's not obvious to me that every packet put
into a network must take the shortest path through the network.
For example, suppose that the load between point A and point B turns out to
be heavier than what the links on the shortest path between A&B can
support, and the rest of the load is light. In this case, one can get
greater performance (either throughput or latency) by using additional
paths, even though they are longer. And if you know up front that some
packets can be delayed longer than others, you even have a heuristic for
picking longer vs. shorter routes.
For all A, B pairs, there will be some in the hierachical decomposition
that will not be able to use hierarchical addresses to discover a set of
routes that do particularly well in spreading the traffic along disjoint,
non-interfering routes. Or so I would think. Not a proof.
All I mean to point out by this is that hierarchy of addresses is
intimately tied with other assumptions about what routing will be used, and
what the load looks like (as well as what I really care about, which is
that topology changes end up being reflected in application-visible
addresses, and thus end up pushing the problem onto the higher levels which
have no better way to deal with the problems.)
To me, wireless and nomadic networks matter, because at some point, we will
likely have many more nodes of that kind than there will be of fixed
kind. Thus today's routing problems don't have anywhere near the scaling
issues that wireless and nomadic ones do.
More information about the end2end-interest