Alime.232
net.lan
utzoo!decvax!harpo!floyd!vax135!lime!martin
Fri Apr 23 09:26:58 1982
Comments on CSMA vs token passing
From: lime!houxe!houxi!ihnss!mhtsa!mhuxt!eagle!karn
typical coaxial cable is .67c, not .3c, although it can be higher
with different impedances and dielectrics.
Token passing has MORE problems in long-propagation-time
environments, not fewer. Just passing the token around the network
can waste a lot of time if there are many nodes in the network.
I don't think anyone can argue that one protocol is inherently
better than others; most have their strengths and weaknesses.
CSMA works better when there are a large number of nodes, each with
relatively little traffic (e.g., terminals); token passing works
better when the traffic builds up (e.g., computer-to-computer file
transfers).
For internal BTL applications, however, you almost have to admit that
ANYTHING is better than the absurd practice of using 212 modems
and voice PBXs for data communications. This makes about as much
sense as IBM trying to switch voice with one of their front-end terminal
processors. Its getting to the point where I don't CARE what is used
to replace the Dimensions; I'd just like to see it soon!
Phil Karn
MH 3C-339
x6646
-----------------------------------------------------------------
gopher://quux.org/ conversion by John Goerzen <
[email protected]>
of
http://communication.ucsd.edu/A-News/
This Usenet Oldnews Archive
article may be copied and distributed freely, provided:
1. There is no money collected for the text(s) of the articles.
2. The following notice remains appended to each copy:
The Usenet Oldnews Archive: Compilation Copyright (C) 1981, 1996
Bruce Jones, Henry Spencer, David Wiseman.
