Re: Bandwidth

Andrew Daviel (
Mon, 5 Jan 1998 12:20:22 -0800 (PST)

Date: Mon, 5 Jan 1998 12:20:22 -0800 (PST)
From: Andrew Daviel <>
To: "Daniel S. Riley" <>
cc: James Green <>,
In-Reply-To: <>
Message-ID: <>
Subject: Re: Bandwidth

On 25 Nov 1997, Daniel S. Riley wrote:

> James Green <> writes:
> But what you are proposing is a cache--the routers (or whatever) have
> to store the pages they are going to serve, and that makes them
> caches.  If we forget caches, we have to forget your proposal too.
> See <URL:> for information on the work on
> national and international cache hierarchies.
> See <URL:> for
> what Cisco has been doing.

I know this is probably dead ... just going through some unread
email here ...

Re. Cisco, as I recall they have been working on a transparent
proxy cache - your average user doesn't realize he's using one since port
80 is intercepted. I don't think it hooks up with Squid-style ICP, though
there is some other point-to-point sharing mechanism I think.

There's the possibility I think of big organisations running
a geographically diverse web farm, so that the DNS/routers 
would find the closest alias, or having a homepage which does
a redirect to the closest mirror (at least, it's relatively simple to
redirect,.edu etc. domains such as .uk, .it etc.) - trying to encourage cache; there
seems to be much more interest in places where the trans-ocean bandwidth
is scarce or expensive - Japan, Australia, etc. - and the academic 
organisations are more supportive than commercial ISPs.

I'm a bit out-of-touch with active HTML etc., though I believe many such
applications are cache-unfriendly.
I'd like to see multicast more widely deployed and used more instead of
real-time video/audio streams; stored realaudio etc. files are 
I think cacheable and better than .wav,.avi etc. anyhow - one can
quit before the end

Andrew Daviel
Vancouver Webpages & TRIUMF