[8752] in Commercialization & Privatization of the Internet
Re: Draft paper available: Prioritized queuing
daemon@ATHENA.MIT.EDU (William Allen Simpson)
Fri Dec 3 00:30:45 1993
Date: Thu, 2 Dec 93 23:21:11 EDT
From: "William Allen Simpson" <bill.simpson@um.cc.umich.edu>
To: com-priv@psi.com
Cc: net-acct@rice.edu
Reply-To: bsimpson@morningstar.com
> From: hwb@upeksa.sdsc.edu (Hans-Werner Braun)
> Our proposal has three elements. First, network routers would queue
> incoming packets by IP Precedence value instead of the customary
> single-threaded FIFO. Second, users and their applications would
> voluntarily use different and appropriate precedence values in their
> outgoing transmissions according to some defined criteria. Third,
> network service providers may monitor the precedence levels of traffic
> entering their network, and use some mechanism such as a quota system
> to discourage users from setting high precedence values on all their
> traffic. All three elements can be implemented gradually and unevenly
> across the Internet infrastructure, providing a smooth transition path
> from the present system. The experience we gain from an implementation
> will furthermore provide a valuable knowledge base from which to
> develop sound accounting and billing mechanisms and policies in the
> future.
>
Goodness, you _are_ behind the times. Every router I've worked on in
the past 4 years has had feature (1).
Most of them also have fair-queuing on a source to outgoing-interface
basis, giving you feature (3).
Not much you can do to change applications to get feature (2). The
criteria are already defined in some RFC, hard to remember which.
But if the application didn't do it, you're stuck.
Actually, the priority can be shifted 1 left, and if it is a telnet
session, 1 is added to the priority. Messy layering, but useful. Thank
Phil Karn for the innovation, it's been in KA9Q for years.
Bill.Simpson@um.cc.umich.edu
