The IP over ATM Mailing List Archive by date

[Gary Hanson <gary@kentrox.com>]

On Wed, 18 Oct 1995, Grenville Armitage wrote:

> 
> Gary,
> 	[..]
> >>Just to avoid possible confusion, I would like to point out that, as with
> >>the VPI and VCI values, a MID value is only of local significance on the
> >>link on which it is used, and has no end-to-end significance whatsoever.
> 
> This is an interesting assertion. The MID is an AAL3/4 level identifier,
> and as such _is_ 'end to end' (where we are talking about the
> ends of a cell flow terminated by AALs). This quite definately gives
> them a broader reach than individual VPI/VCI values (which may change
> along the path of a VCC), or just VPI values (which may change along
> the path of a VPC).

Sorry, er, I should have clarified that I was referring to the SMDS/CBDS
case with my assertion about the MID value's significance, and that by
"link" I was referring to the subscriber-network interface (SNI) link.
I certainly did not mean to ADD confusion here... :-)

Actually, what SMDS/CBDS switches do between themselves has always been
an interesting issue.  My understanding has been that the earliest SMDS 
switches actually reassembled the SMDS L2_PDUs (i.e., the SMDS "cells"),
and then forwarded the SMDS L3_PDU to the next or the final egress SMDS
switch as full frames, there to be re-segmented with new MIDs on the
egress SNI.  I had assumed that this was merely the easiest way to evolve
the infrastructure, and that it was a temporary transport solution.  I
had always hoped that SMDS switches would evolve to use cell-switching
technology, but I am not privy to knowledge about SMDS current practice.

You are right, of course, in AAL3/4 the MID is truly end-to-end, for use
in multiplexing multiple cell flows over the VCC.

> puzzled,
> gja
> 

Regards,
Gary