The IP over ATM Mailing List Archive by date

Cell Relay Retreat>List Archive>month:1996-Mar> msg00264



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

[Q] Suggest new protocol providing QoS...

  • From: Jon Crowcroft <J.Crowcroft@cs.ucl.ac.uk>
  • Date: Sun, 31 Mar 1996 09:54:52 +0100
  • Cc: fred@cisco.com (Fred Baker), ip-atm@matmos.hpl.hp.com

 >I'm not asking if IPs important to us; I'm asking if IP is important (enough)

good question

most market analysts now admit, though, that VOD is not likely to be i
 great demand

(and anyhow if it is, the DAVI people have a far more cost effect set
of technologies for delivrring it to existing POTS/Cable subscribers,
which is a set of users that still outstrips Internet sbscribers by at
least 1 order of magnitude)
 

but my point on encapsulatint#g ATM in new NNI headers (suggestion for
supporting IP multicast model scalably), is that if ATM can't do it,
it is a good example of why IP is so succesful - the whoele Mbone
switched from using loose source routing to IP in IP in the space pf a
few weeks, and the whole Mbone is bigger than nay dfeployed ATM net
(couting end sytems....easily...)

if atm must change, it must change to become more flexible...or, as a
manager of network change, and noting that this is the most complex
and expensive element of management in any network, i would reject 
it as to inflexible....

 jon