The IP Over NBMA (ION) Archive

Cell Relay Retreat>ION Archive>month:1996-Jul> msg00097



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

Yet another multicast scalability problem

  • From: Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp>
  • Date: Wed, 24 Jul 96 10:45:36 JST
  • Cc: mohta%necom830.hpcl.titech.ac.jp%3COM@smtp1.isd.3com.com (Paul Koning/US/3Com), 0006078043%mcimail.com%3COM@smtp1.isd.3com.com (Paul Koning/US/3Com), mohta%necom830.hpcl.titech.ac.jp%3COM@smtp1.isd.3com.com (Paul Koning/US/3Com), ion%nexen.com%3COM@smtp1.isd.3com.com (Paul Koning/US/3Com), suzuki%nal.ntt.jp%3COM@smtp1.isd.3com.com (Paul Koning/US/3C i)

> >> Is control over loss
> >> probability not useful to IP?
> >
> >No, it is not for best effort multicast IP.

> The most that could reasonably be said is that control of loss probability
> is not as critical for a particular protocol.  But all protocols benefit from 
> low
> loss channels.

The problem is that, with multicast, the meaning of "low loss" is
fuzzy.

If the lowest available bandwidth of the entire multicast tree
is 64Kbps and the lowest available bandwidth of part of the
multicast tree, which covers 99% of receivers, is 6Mbps,
which is the proper bps for low loss communication?

							Masataka Ohta