From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ig0-x233.google.com (mail-ig0-x233.google.com [IPv6:2607:f8b0:4001:c05::233]) (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 4E57821F307 for ; Tue, 27 May 2014 21:59:23 -0700 (PDT) Received: by mail-ig0-f179.google.com with SMTP id hn18so2023932igb.0 for ; Tue, 27 May 2014 21:59:22 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type; bh=yaXIzUy0lorglMfQU1x4uz9cCfVLKf98QzjXgAJTRpg=; b=oCsBpmlr6OY7RG2hMR8cdSPLFo2N6sbCsuQLWKQUmD4Rg+Kxy8xns3REJRqeqNza0g 8OVroTB5ZvsP6Mc/w3l0N1y9/KWZmYoxB85a+keLjL1tkGl/MRZfxE1JgGnjO98uxQnw M7DErd2hbP7FR1HM+qzDVdLTkf5TX88YkEcHYY7P5fqU+hZ3FjFhXR91FLMcz/JyaL9J ZkVVMTf7E+Upa1pKEAlUBbu9W2GwfTj53AfV4OT2bjABHNBNq4tjaZsvPVE/b4MwhEPd F5jgGdtwUZToq+oQ7zGiv6sscsljl39Ap+yZmumLaHKhPabiImEYzVPzikF1tPBQSuIB kczw== MIME-Version: 1.0 X-Received: by 10.50.43.225 with SMTP id z1mr38882805igl.29.1401253162495; Tue, 27 May 2014 21:59:22 -0700 (PDT) Received: by 10.64.59.165 with HTTP; Tue, 27 May 2014 21:59:22 -0700 (PDT) Date: Tue, 27 May 2014 21:59:22 -0700 Message-ID: From: Aaron Wood To: cerowrt-devel Content-Type: multipart/alternative; boundary=089e010d8dd6fdb53204fa6eac2d Subject: [Cerowrt-devel] pie vs. fq - no contest X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 28 May 2014 04:59:23 -0000 --089e010d8dd6fdb53204fa6eac2d Content-Type: text/plain; charset=UTF-8 http://burntchrome.blogspot.com/2014/05/i-dont-like-this-pie-more-bufferbloat.html PIE certain is better than nothing. But it's no fq_codel. All I varied between the two is the queuing discipline, with the PIE target (both directions) set to 16ms. not impressed. But, far better than pie did when I was in Paris, on a 16Mbps down, 1Mbps up ADSL2+ link: http://burntchrome.blogspot.com/2014/01/comparing-queue-disciplines-in-cerowrt.html I think those results were either the default target being wrong, or the highly asynchronous and highly limited setup being ill-suited to PIE. Which is concerning, as PIE is going to be deployed in 30/1 cable setups (like the Numericable.fr setup I had in my first apartment in Paris). -Aaron --089e010d8dd6fdb53204fa6eac2d Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
http://burntchrome.blogspot.com/2014/0= 5/i-dont-like-this-pie-more-bufferbloat.html

PIE= certain is better than nothing. =C2=A0But it's no fq_codel.

All I varied between the two is the queuing discipline,= with the PIE target (both directions) set to 16ms.

not impressed.

But, far better than pie did when= I was in Paris, on a 16Mbps down, 1Mbps up ADSL2+ link:


I think those results were either the default target being wrong, or t= he highly asynchronous and highly limited setup being ill-suited to PIE. = =C2=A0Which is concerning, as PIE is going to be deployed in 30/1 cable set= ups (like the Numericable.fr setup I had in my first apartment in Paris).

-Aaron
--089e010d8dd6fdb53204fa6eac2d--