The IP Over NBMA (ION) Archive

[Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp>]

Brian;

> Correct I think. NHRP would need a heuristic to "freeze" it
> part way through the cloud to avoid this pessimisation.

My proof uses nothing NHRP specific.

It is simply proven that cut-through doesn't scale.

And, gja;

> I've been working on a short document discussing this very issue,
> which I hope to release as an informational I-D asap. In it
> I look at the requirements placed on a hypothetical multicast
> cut-through protocol, and conclude that any system that supports
> multicast cut-throughs amongst participating clusters is no
> different (and indeed, more complex) than simply placing all
> hosts across your ATM cloud into one big cluster.

Further clarification of my point above is, of course, welcome
as long as the memo is not so thick.

> Of course, since every good protocol needs a good acronym, I
> called it the Very Extensive Non-Unicast Service - VENUS.

A problem is that Venus has thick atomosphere.

						Masataka Ohta