The IP over ATM Mailing List Archive by date

[Curtis Villamizar <curtis@ans.net>]

In message <199505170533.OAA10255@necom830.cc.titech.ac.jp>, Masataka Ohta writ
es:
> > > If you insists that packet-wise relaying is enough, you should let
> > > the IPATM WG die, instead.
> > 
> > Why.
> 
> Because LANE has accomplished exactly what you need in a way
> a lot more suitable than the classical model.
> 
> > We expect to be buying ATM service and equipment and want IP
> > over ATM to work (very well if possible).
> 
> Very well means "with QoS of ATM" and it is possible.

That is incorrect.  You are insisting that one approach to solving a
problem is the requirement.  The requirement is that both elastic
traffic and real time traffic are well accommodated and handled
efficiently.  "QoS of ATM" may be one approach, but better solutions
might exist.

> Otherwise, you can't say seamless LAN-WAN connection, anymore.
> 
> > I don't see "being
> > connected oriented" as an explicit goal and see no reason to abandon
> > IP over ATM for that reason.
> 
> The goal, of course, is "connectionless at the Internetwork layer,
> but being able to be reservation oriented at the link and the transport
> layer".
> 
> 							Masataka Ohta

The goal is not to "be connectionless" or to "be reservation oriented".
The goal is to accomodate both elastic traffic and real time traffic.

Can we just agree that we have differing views as to which approaches
to solving the problem will be the most effective and end this thread?

Curtis