From owner-freebsd-net@FreeBSD.ORG Mon Dec 10 19:04:04 2007 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 400BA16A419 for ; Mon, 10 Dec 2007 19:04:04 +0000 (UTC) (envelope-from rihad@mail.ru) Received: from mx28.mail.ru (mx28.mail.ru [194.67.23.67]) by mx1.freebsd.org (Postfix) with ESMTP id F0FE513C458 for ; Mon, 10 Dec 2007 19:04:03 +0000 (UTC) (envelope-from rihad@mail.ru) Received: from mx30.mail.ru (mx30.mail.ru [194.67.23.238]) by mx28.mail.ru (mPOP.Fallback_MX) with ESMTP id B847BCB29A for ; Mon, 10 Dec 2007 19:57:31 +0300 (MSK) Received: from [217.25.20.94] (port=53047 helo=[217.25.20.94]) by mx30.mail.ru with esmtp id 1J1lwf-0000Is-00 for freebsd-net@freebsd.org; Mon, 10 Dec 2007 19:57:29 +0300 Message-ID: <475D6FD7.2000500@mail.ru> Date: Mon, 10 Dec 2007 20:56:55 +0400 From: rihad User-Agent: Icedove 1.5.0.14pre (X11/20071018) MIME-Version: 1.0 To: freebsd-net@freebsd.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Pipe queues X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Dec 2007 19:04:04 -0000 Hi, I'm having a hard time to understand what pipe queues are with respect to bandwidth limitation. ipfw(8) and dummynet(4) manuals didn't help me much. How does dummynet's traffic shaping function? Why does it need a queue at all? As each connection's queue takes some memory, can I just disable the queueing, or set its size to 1 slot (1500 bytes)? What if I set it to 10000 slots, why should it wait to fill the queue before starting to send stuff out the network interface? As packets arrive, AFAIK they are stored in mbufs in system memroy. Do pipe queues also use mbufs? I might be servicing hundreds of pipes simultaneously, this is why I'm concerned about memory use. Thank you for any insights.