From: Eugene Leitl (Eugene.Leitl@lrz.uni-muenchen.de)
Date: Tue Jul 10 2001 - 04:58:15 MDT
On Mon, 9 Jul 2001, Brian D Williams wrote:
> Okay, so you plan to use encapsulation to accomodate IP.
Yup.
> I still don't see the routing part, you're not using a distance-
Little wonder, as I don't have a specific implementation, just a set of
fuzzy ideas.
> vector (Bellman-Ford) nor a link-state (OSPF or Dijkstra) protocol or
> even a hybrid (EIGRP, BGP), essentially it seems to be nothing more
> than a simple broadcast which would eat up your bandwidth, not to
> mention propagation delays, and what about loops?
The nodes are labeled with geodetic (presumably polar, as matter tends to
self-organize in hierarchical orbital assemblies) coordinates. You know
which coordinates your direct neighbours have (and maybe their neighbours,
if your wiring deviates too far from orthogonal).
The dumb approach would be just forward (probably fixed-sized) packets
through the link to the node with the coordinates geometrically nearer to
the target (here's your proximity metric, it's just an n-dimensional
Bresenham). Real world implementations would accomodate for
unorthogonality of realworld networks, and preferably use less loaded
links.
-- Eugen* Leitl leitl
______________________________________________________________
ICBMTO : N48 10'07'' E011 33'53'' http://www.lrz.de/~ui22204
57F9CFD3: ED90 0433 EB74 E4A9 537F CFF5 86E7 629B 57F9 CFD3
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:08:37 MST