Re: Real-time distributed Web search (Gnutella knockoff?)

Date view Thread view Subject view Author view

From: Adam L. Beberg (
Date: Fri Jun 30 2000 - 19:28:30 PDT

On Fri, 30 Jun 2000, Meltsner, Kenneth wrote:

> There was some good work on this by the Hyperwave folks in Austria.
> Basically, in order to assure complete propagation of update messages
> without requiring nasty full-blown broadcasts, they send the update messages
> to a couple of servers, which send them to a couple of servers each, and so
> on. Gross simplification, of course, but they showed both the math and some
> practical experiments in the n=thousands range that this works well. Don't
> know if their ideas scale into the hundreds of thousands or millions of
> computers, though, but it did handle the "natural" hierarchies imposed by
> network-network connectivity as well.

Yeap, that's the basic idea behind a dynamic hierarchy, you have parts
that are closer then other parts, so you talk to them. It's kinda the
6-degress of separation thing, which I believe they figured out for the
internet was ~30. Reading section 6.3 and knowing the size of the
net, I'd say they don't scale well, but then it's an expensive

It's not a broadcast problem to begin with, it's an inform-a-list
problem. That's what refer logs are for, getting the list. I've actually
done this manually when I moved my HL site to it's own domain long
ago. Get the list, email people, take a nap. No broadcast required.

- Adam L. Beberg
  Mithral Communications & Design, Inc.
  The Cosm Project - -

Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Fri Jun 30 2000 - 19:32:07 PDT