The Routing Over Large Clouds Mailing List Archive by date

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

> Ohta-san,
> 
> > "fewer VCs required on the ATM network"?
> > 
> > Is it a so infamous multipoint-to-point cell-relay hype?
> > 
> > Or, are you suggesting to use packet relay and lose all the merit,
> > IF ANY, of cell-relaying architecture? But, then, what "if it stays
> > at the ATM layer" means?
> 
> ?? I'm sorry I have no idea what you mean...

I mean? I'm asking what your statement

	> > "fewer VCs required on the ATM network"?

means.

> > BTW, the more intermediate routers you skip, the more VCs you
> > need to connect to distant routers. The number of VC grows
> > exponentially to the number of skipped routers.
> 
> NO. It strongly depends on how many hosts and routers on the network
> and how to communicate each other.

It strongly depends?

But, it's you who claimed

	fewer VCs required on the ATM network

without stating the dependency.

If you now say "it depends", your statements was hype.

> Anyway, I had no intention to argue about the situation where you are
> thinking of (back ground, a specific technology...).

In which situation I am thinking of?

I'm asking in which situation your statement can be true.

What is your situation?

> What I meant in
> my previous mail was about general NHRP spec. (^o^)

As you said "It strongly depends on how...", you can't say "general
NHRP spec.".

Well, if you want to say that NHRP works over a small cloud, yes,
you are right.

							Masataka Ohta