From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-we0-x236.google.com (mail-we0-x236.google.com [IPv6:2a00:1450:400c:c03::236]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id DCC4F21F1FC for ; Thu, 20 Mar 2014 13:23:09 -0700 (PDT) Received: by mail-we0-f182.google.com with SMTP id p61so995471wes.27 for ; Thu, 20 Mar 2014 13:23:08 -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:content-transfer-encoding; bh=Q+Hx+O4iDs0/eHBg+6zAiARHuduYy9PnkTWVhhFapGc=; b=Nra1dQXnz9ypaqMeKgO8KFDzjluYwHqoW7D0fT+OAVFVstVKQJsiD0uzMsBzfyk/JA R4dpnNckfO2IYR9XxEY21k8N+sfrb5P3S/C11s/ViEQ+NkOR44VikPo4XhNgxdDuF4Ug 55zPSuLdioJYbkAp564tU7QXk2r/AWHyTox86OJuReVBlc96ijIxfZqNiSUoqaIVqoam n8N0Kb+K9BMrY/aYSE8IFSclbth08dZRgFqzLoB8LalPWAOXxHcfvUB6Qy7MX9iAijk2 LT0uEMAFog6ApN2XKFvhATAJ8Li00Z6li/cbI+Op5uT+2EsbMiqJLQHuiPeXZi6JbBos P+Aw== MIME-Version: 1.0 X-Received: by 10.180.189.139 with SMTP id gi11mr4915946wic.53.1395346988101; Thu, 20 Mar 2014 13:23:08 -0700 (PDT) Received: by 10.216.8.1 with HTTP; Thu, 20 Mar 2014 13:23:08 -0700 (PDT) In-Reply-To: References: <20140318145221.GA31327@sesse.net> <07BD4518-2A7E-4F43-8978-791E3B2BDA2A@cisco.com> <87eh1wc05c.fsf@toke.dk> <87a9ckbz1q.fsf@toke.dk> Date: Thu, 20 Mar 2014 13:23:08 -0700 Message-ID: From: Dave Taht To: Aaron Wood Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: renaud sallantin , bloat Subject: Re: [Bloat] AQM creeping into L2 equipment X-BeenThere: bloat@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: General list for discussing Bufferbloat List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Mar 2014 20:23:10 -0000 On Thu, Mar 20, 2014 at 12:34 PM, Aaron Wood wrote: >> however what we are probably seeing with the measurement flows is >> slow start causing a whole bunch of packets to be lost in a bunch. > > > That would line up with the timing, and the periodic drops that I see in = the > flows when using Toke's newer wrapper (and netperf head), which attempt t= o > work around the failing UDP timing flows. Well there is some good work in linux 3.14 and beyond, and there was also some interesting work on "initial spreading" presented at ietf. Hopefully patches for this will be available soon. http://tools.ietf.org/html/draft-sallantin-iccrg-initial-spreading-00 I would certainly like to be able to sanely measure the impact of hundreds or thousands of flows in slow start, rather than/in addition to 8 = flows in congestion avoidance. > > -Aaron --=20 Dave T=E4ht Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.= html