From owner-freebsd-net@FreeBSD.ORG Wed May 9 01:41:39 2012 Return-Path: Delivered-To: net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 2DB5B106566B for ; Wed, 9 May 2012 01:41:38 +0000 (UTC) (envelope-from kob6558@gmail.com) Received: from mail-pb0-f54.google.com (mail-pb0-f54.google.com [209.85.160.54]) by mx1.freebsd.org (Postfix) with ESMTP id BD7738FC15 for ; Wed, 9 May 2012 01:41:38 +0000 (UTC) Received: by pbbro2 with SMTP id ro2so9898520pbb.13 for ; Tue, 08 May 2012 18:41:38 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=U++I917cHzMJiRxErVYHUPQ7G4RytlL12OqjEEJY4D8=; b=sA60aa7V7ZtRu8sBh95bZz5E1p9R/rsGzMjkIiDgzViFVF31lMBum6UXu/V2Es8xvb +peXFp86UrIaNO35i9WzDZUL6kIi4E12+cy424hCOPr2hZehLJUwKjKmwGd7TefuRBiZ 2K2+G0C7gKzG6zMjnFqqk/YWbrZOwbNb5rrYUvg506em5PrV5N3/WXsjbbQiN6j6G/zt Vt69iffemk5VVeoWMyJpAn55DUjgyLz7spKB3Z+l+V4xNMo5ClgwVcoZFVySFStdmCwu CRYd+uptcU598bNmRdS/JsnbWoXc65TZlZZwDkh2RmeBjpVH/2TbKCg4l2csZEJ8Drs7 U+zQ== MIME-Version: 1.0 Received: by 10.68.225.74 with SMTP id ri10mr228914pbc.40.1336527698056; Tue, 08 May 2012 18:41:38 -0700 (PDT) Received: by 10.68.130.102 with HTTP; Tue, 8 May 2012 18:41:38 -0700 (PDT) In-Reply-To: <20120508190343.GA72070@onelab2.iet.unipi.it> References: <20120508190343.GA72070@onelab2.iet.unipi.it> Date: Tue, 8 May 2012 18:41:38 -0700 Message-ID: From: Kevin Oberman To: Luigi Rizzo Content-Type: text/plain; charset=ISO-8859-1 Cc: net@freebsd.org Subject: Re: Controlling queue delays 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: Wed, 09 May 2012 01:41:39 -0000 On Tue, May 8, 2012 at 12:03 PM, Luigi Rizzo wrote: > Here is a recent paper that revisits active queue management. > > http://queue.acm.org/detail.cfm?id=2209336 > > might be interesting for those involved in TCP cong.control, > traffic schedulers (ok, that includes me!) and the network > stack in general I started looking at this yesterday and it looks like something that should be a requirement for any system as it should fix, not work-around, the issues of buffer-bloat that is becoming a major issue on more an more networks, both for business, but especially homes. This assumes that it actually works as advertized, but the authors are unlikely to have published this without thorough analysis and testing. They are, after all, among the leading TCP experts in the world. -- R. Kevin Oberman, Network Engineer E-mail: kob6558@gmail.com