From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-io0-x22f.google.com (mail-io0-x22f.google.com [IPv6:2607:f8b0:4001:c06::22f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 464643B2A4 for ; Thu, 6 Apr 2017 22:52:10 -0400 (EDT) Received: by mail-io0-x22f.google.com with SMTP id b140so42123050iof.1 for ; Thu, 06 Apr 2017 19:52:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=bitamins-net.20150623.gappssmtp.com; s=20150623; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=3udx6iBdfWGUO68BNWCMAoz99paK5/hJVrK/Suz6uUk=; b=MaJBv2ZebUaHtTmGjoOtUKjb/uBi0MEHS15zcEae0rdNmQOTZLJVQ4vpWVGHx7W1SZ jOw/m8b28l6K5RatJ2uoIeIjS86+qTYBIgoCHvjmykRqXZhmEeNnrLh7O31F+VcoDvZa 9I7UaOmnwUTgb2l+fgULnOT8oPa7f4GDZywXYdAQwEjMlrCf/dvJVBVriDRcMo6NBmhr oM7GpVLny07TIBQmq9vy2YL0iwp86TJ3ZTco34vR3X0G/u3+Lhc0rOeWKgbYz5Zaej5f kNS6uvCB/mUkPcTb0zzpVMnDdj5ABX7t0ROBQiTi02vXhuRWlwyDwq7NbdL7rmL8RSyl DObw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=3udx6iBdfWGUO68BNWCMAoz99paK5/hJVrK/Suz6uUk=; b=fBH3cN2ELiPPCVMlZUo5m3OlV7VU40PrU7eRA9tExNiqkat3VklN52pxmf6/qQX6Df 6UZBPeOT6NiTda13x8FPfnrGq2PbdCyWDo7utuBWaczWiTXSfwvJ+n6GonlwdYdCSg0O MXoENDnSJFAo94uaeairjTmcBUdFNsfulMMyGtui+4FPMZzizcY4TgAd8hp+HJc8ddkl LpJDdw028L8nB8DceTQrWJl49jNMwk5bejnbhncU9pHKIvzgFJ0hXQrsrHeFe0rq8V7/ ZsVTiwwRATEU0Woi4csGuJ53sxm1lafS+S6dfqi004D43MPpGK8JgqBhRCA9qyu+1ICT 92Wg== X-Gm-Message-State: AFeK/H2ei10H1dKfvnONqrXvw/Oka5/Lu/WqtqF20LZWBhpEr5493qDIpwTSXv5YfLev4jV4azy6cL56y562+A== X-Received: by 10.107.137.103 with SMTP id l100mr34770420iod.215.1491533529713; Thu, 06 Apr 2017 19:52:09 -0700 (PDT) MIME-Version: 1.0 References: <3FFFF4B7-6DF7-4F7B-8F2D-5E22E27769FB@gmail.com> In-Reply-To: <3FFFF4B7-6DF7-4F7B-8F2D-5E22E27769FB@gmail.com> From: "Luis E. Garcia" Date: Fri, 07 Apr 2017 02:51:59 +0000 Message-ID: To: Jonathan Morton Cc: cake@lists.bufferbloat.net Content-Type: multipart/alternative; boundary=001a113ecff635ad01054c8ab678 Subject: Re: [Cake] A quick question about FQ_CoDel vs Cake 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: Fri, 07 Apr 2017 02:52:10 -0000 --001a113ecff635ad01054c8ab678 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Thanks Jonathan, With this explanation it makes sense. Regards, Luis On Thu, Apr 6, 2017 at 11:06 AM Jonathan Morton wrote: > > > On 6 Apr, 2017, at 16:51, Luis E. Garcia wrote: > > > > I've been doing some testing of Cake on LEDE (WD MyNet 750) and on > EdgeOS (Ubiquity ERPoe). One big question that I have is why does Cake ha= ve > a higher/better average throughput than FQ_CoDel? The graph seems a bit > smoother through the speed test. > > > > The test are against a 10down/2up Mbps connection from a local provider= . > > The main difference that=E2=80=99s probably responsible for this is Cake= =E2=80=99s > integrated deficit-mode shaper, which is more accurate on short timescale= s > than the more typical token-bucket shaper that fq_codel is used with. > > There=E2=80=99s also some difference in the Codel implementation which mi= ght or > might not be relevant, specifically in the calculation of =E2=80=9Ccount= =E2=80=9D after a > relatively brief exit from dropping state. > > - Jonathan Morton > > --001a113ecff635ad01054c8ab678 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Thanks Jonathan,
With this explanation it makes sense.
=

Regards,
Luis

On Thu, Apr 6, 2017 at 11:06 AM Jonathan Morton <chromatix99@gmail.com> wrote:
<= /div>

> On 6 Apr, 2017, at 16:51, Luis E. Garcia <luis@bitamins.net>= wrote:
>
> I've been doing some testing of Cake on LEDE (WD MyNet 750) and on= EdgeOS (Ubiquity ERPoe). One big question that I have is why does Cake hav= e a higher/better average throughput than FQ_CoDel? The graph seems a bit s= moother through the speed test.
>
> The test are against a 10down/2up Mbps connection from a local provide= r.

The main difference that=E2=80=99s probably responsible for this is Cake=E2= =80=99s integrated deficit-mode shaper, which is more accurate on short tim= escales than the more typical token-bucket shaper that fq_codel is used wit= h.

There=E2=80=99s also some difference in the Codel implementation which migh= t or might not be relevant, specifically in the calculation of =E2=80=9Ccou= nt=E2=80=9D after a relatively brief exit from dropping state.

=C2=A0- Jonathan Morton

--001a113ecff635ad01054c8ab678--