From owner-freebsd-isp Sun Jun 8 07:48:41 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id HAA14932 for isp-outgoing; Sun, 8 Jun 1997 07:48:41 -0700 (PDT) Received: from stingray.ivision.co.uk (stingray.ivision.co.uk [194.154.62.72]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id HAA14927 for ; Sun, 8 Jun 1997 07:48:33 -0700 (PDT) Received: from stingray.ivision.co.uk [194.154.62.72] by stingray.ivision.co.uk with smtp (Exim 1.62 #1) id 0wajEq-0000yU-00; Sun, 8 Jun 1997 15:46:36 +0100 Date: Sun, 8 Jun 1997 15:46:36 +0100 (BST) From: Manar Hussain Reply-To: Manar Hussain To: Luigi Rizzo cc: dennis , freebsd-isp@freebsd.org Subject: Re: ETinc's Bandwidth limiter In-Reply-To: <199706072008.WAA03112@labinfo.iet.unipi.it> Message-ID: Organisation: Internet Vision MIME-Version: 1.0 MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-isp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >> There is no "fair routing" in a web farm unless everyone pays the same >> price, which is ridiculous. Charge based on their bandwidth access >> capability.. >> and with the bandwidth manager there is not accounting headaches 'cause >> they cant get more than they pay for. > >Dennis, this sounds like an overstatement. Fair does not necessarily >mean 'all equals', there can be different weights for different >users depending on how much they pay for, and the fairness is in >making everyone get what he pays for. Hard limiting the bw for >each user as you seem to suggest prevents eveyone from taking >advantage of statistical multiplexing, which, given the burstiness >of network traffic, is very rewarding for all. Exactly. There are a whole host of ways I can fairly happily limit each hosts bandwidth if I'm not bothered by these limits being "soft". The aim of the game is to be able to confidently offer a minimum level of service (which they can specify and thus pay for) but let people make more out of it if they can. Manar