From owner-freebsd-isp Mon Apr 6 15:15:42 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA22977 for freebsd-isp-outgoing; Mon, 6 Apr 1998 15:15:42 -0700 (PDT) (envelope-from owner-freebsd-isp@FreeBSD.ORG) Received: from mail.keyworld.net (root@mail.keyworld.net [194.21.164.2]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA22970 for ; Mon, 6 Apr 1998 15:15:30 -0700 (PDT) (envelope-from chrism@keyworld.net) Received: from chrism (ppp71.keyworld.net [194.21.164.134]) by mail.keyworld.net (8.8.5/8.8.4) with ESMTP id AAA31376; Tue, 7 Apr 1998 00:11:12 +0200 Message-Id: <199804062211.AAA31376@mail.keyworld.net> From: "Christopher Martin at Home" To: "Jim Shankland" , , , Subject: Re: Bandwidth limiter for services? Date: Tue, 7 Apr 1998 00:14:17 +0200 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1155 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: owner-freebsd-isp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I'm sure Dennis will be heard from, and will bring his customary > diplomatic balm to bear on this matter, but: note that this is > no different from what happens when packets leave your 10- or > 100-Megabit-connected workstation and are routed over a T-1 or slower > WAN line. TCP already deals with this, using slow start, congestion > avoidance, etc. to deduce the amount of bandwidth actually available > to it, and throttle itself back accordingly. Your notion that > a bandwidth limiter like ET's will cause increased retransmissions > is unfounded. > > Jim Shankland > Flying Fox Computer Systems, Inc. How about setting throttles on the hosts from the start in order to avoid interim congestion prior to the normal TCP cutback... Do anyone of you live in places where a T1 costs USD 500,000 per annum? -- Christopher Martin BDM -KeyWORLD http://www.keyworld.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message