From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-lf0-x231.google.com (mail-lf0-x231.google.com [IPv6:2a00:1450:4010:c07::231]) by lists.bufferbloat.net (Postfix) with ESMTPS id CF1083C9F3 for ; Mon, 18 Jan 2016 04:37:40 -0500 (EST) Received: by mail-lf0-x231.google.com with SMTP id m198so168384731lfm.0 for ; Mon, 18 Jan 2016 01:37:40 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=content-type:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=ZXe6ugvEbSiZiYTgNy9UUufYqhpUYMQckLTBg5ZcwVo=; b=RaONJZb0seO+XqH+ib/21BlYLjPOCbcbOPTrp7Pk8RnFu8zkCU0yHTz/Q/3F/jysAS UiHi5ZdzAGdjkBsrkqOSr2u9AxesHQ4CPCAcCP7KMVaDnXEmlyiGdU41M0CWX/bLo7k1 xW4tUytdtvEzZpKCgGxk/W/BtFMn3/o49IZ/9eydnZNyqvaGvxaWEz87cGlf53VKYa00 phH95T8edzZkiTtMKBp35h/gPkXO0RndcjObTHzZxS8CcAwlu1aSUYVtFz+IQJIOQoe0 cKtTnibXtdemyDe7Ws472rasvxn1JrKmBtmu7WRPd5KHYMYIX0sqcP+bWe/qUcZ+/dtw wS3w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:content-transfer-encoding:message-id:references :to; bh=ZXe6ugvEbSiZiYTgNy9UUufYqhpUYMQckLTBg5ZcwVo=; b=QFAZRskm4pX4e18QevCPSz3tFMrSnBRG3XsuatzqtjLJIF7lXz5BDIzMb+yNGPddlS 116K7vOXDtBknBwOyRLh+CmwcOrcgjXXwkypE169p2qgjf8aqGUYfdO/+3guAsCK5U5X EYXOyTUwuH76lGVUlQ1ROwcqQYcLqnIoNmhEcNd+XqJjc5G1DKwmcAH+CAnJwBY/VQB6 zJ+/Oohiyyg6sGilNsgSTq19+bvB/Is0Tt/HmFktgsR1blW5/IJr+lTohSE6fWFUjyOm pVgZFRUIadr3f+iiVEs5ltoRgUir6zDUUCZjaQzb0iuNCIzb1oM6IbzfDi/yKrn+3Zqd BPkg== X-Gm-Message-State: ALoCoQngR86xsjexoq+lTINVOIJdpYbmVgLxu5yFXnZoAkUhTt/5wRLdzbeao8W08v+u5C+8xJfY3cQqHowC4coa4fEr0AhgeA== X-Received: by 10.25.81.133 with SMTP id f127mr6963136lfb.141.1453109858382; Mon, 18 Jan 2016 01:37:38 -0800 (PST) Received: from [192.168.238.201] (37-33-99-74.bb.dnainternet.fi. [37.33.99.74]) by smtp.gmail.com with ESMTPSA id r63sm3043405lfe.38.2016.01.18.01.37.37 (version=TLS1 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Mon, 18 Jan 2016 01:37:37 -0800 (PST) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 9.2 \(3112\)) From: Jonathan Morton In-Reply-To: <079A86FB-D2A6-4787-BC42-D2E200AD3290@gmx.de> Date: Mon, 18 Jan 2016 11:37:35 +0200 Cc: cake@lists.bufferbloat.net Content-Transfer-Encoding: quoted-printable Message-Id: <72B1E8E5-F42D-4B13-A404-5DE82F8658F9@gmail.com> References: <5693E8FA.4000803@darbyshire-bryant.me.uk> <56941191.1010601@darbyshire-bryant.me.uk> <452D0F47-931B-4412-AC59-C308388AA1E4@gmail.com> <02A10F37-145C-4BF9-B428-BC1BDF700135@gmx.de> <079A86FB-D2A6-4787-BC42-D2E200AD3290@gmx.de> To: moeller0 X-Mailer: Apple Mail (2.3112) Subject: Re: [Cake] triple flow isolation X-BeenThere: cake@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Cake - FQ_codel the next generation List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 18 Jan 2016 09:37:41 -0000 > On 18 Jan, 2016, at 11:21, moeller0 wrote: >=20 > Am I right to assume that dust and src host isolation works with the = same counters but simply ignores one of them? Yes. That=E2=80=99s explicit in the code. > So if all internal hosts talk to one external host, does this scheme = then equal pure per-flow fairness? I am trying to understand how robust = triple-iso is going to be against attempt at shenanigans by unruly = machines on the internal/external networks=E2=80=A6 No. If there is only one host on one side (whichever side that happens = to be), then maintaining per-host fairness for that side is trivial. = The algorithm will instead maintain per-host fairness for the other = side. This derives from the fact that it also maintains per-host = fairness within traffic to each individual host. Per-flow fairness is = also still maintained within individual host-pairs. My measurements show that this aspect of the isolation is not perfect. = There is still some influence from the number of flows, which biases the = actual throughput slightly from ideal per-host fairness. This bias is = however *much* smaller than pure per-flow fairness would be, and I=E2=80=99= ve been unable to come up with a robust way of eliminating it entirely. Hence I think it=E2=80=99s reasonable to simply switch on triple = isolation by default, in the near future. It does approximately the = right thing, without further configuration, in the great majority of = practical cases (that I can think of), and to a greater extent than the = existing =E2=80=9Cflows=E2=80=9D mode does. I think that might also be a good time to overhaul the documentation and = do some other overdue cleanup. - Jonathan Morton