Date: Fri, 30 Sep 94 10:41:13 CDT
From: liberte@ncsa.uiuc.edu (Daniel LaLiberte)
Message-Id: <9409301541.AA01735@void.ncsa.uiuc.edu>
To: Michael.Mealling@oit.gatech.edu, nje@ansa.co.uk
Subject: Multicasting vs Anycasting (was: No "TOP" of the docuverse)
From: Nigel Edwards <nje@ansa.co.uk>
[...]
Client/server multicasting can cause scaling problems in wide-area
networks. It is possible to build protocols in which a client can
send a multicast as single message to m servers on the same LAN
segment. However, if the servers are in different parts of the network
(URN/URC servers will be), m separate messages will be needed to send
a multicast to them (even routers are used to fan-out the
multicast). So if there are n clients you have n*m messages.
The question of which server replies can cause further scaling
problems unless the server which should reply is already known. In
which case why not have the client choose the server and send a
uni-cast message?
I agree that multicasting has problems. I believe anycasting provides
a better solution to the problem where you really only want one of
several servers to respond. Rather than have client applications
decide which server to select, the internetwork can decide based on
proximity, availability, or other criteria. Anycasting is described
in "Host Anycasting Service" <ftp://ds.internic.net/rfc/rfc1546.txt>.
Daniel LaLiberte (liberte@ncsa.uiuc.edu)
National Center for Supercomputing Applications