From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ob0-x22d.google.com (mail-ob0-x22d.google.com [IPv6:2607:f8b0:4003:c01::22d]) (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 3354421F2DB for ; Wed, 18 Mar 2015 14:20:53 -0700 (PDT) Received: by obcjt1 with SMTP id jt1so20452105obc.2 for ; Wed, 18 Mar 2015 14:20:53 -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=UvRvupJZ9+DpGDxboLWHhh5T0CxfVw5BiZhgdAVMZZ0=; b=vnC1se/sNjr1eoPzfVDMfkZ3XwAR4ITGtMp30nOrQNoN+saPGZItNV6kYC1Xj9hpK6 8x5wZ2q0nuzlCMnM2iJiCMDVR04q9tZj6nwVfEUx7/WScT/Xmxolt9sjXN72YwO6jHQd 1uf7i2Wt+z7VF5FmqcqWXU3291KaiS3Nt3TeQaTIB3/NemTu3lBKPo3lSIVGGm7NFewt M2gyxElFEi+02dqycK+bVtup4CzN1yRE4zvw+YHunj0uOf1KpsMTNHWq4PuBLZ5yx0MY ZPsg2Q7dekPcgxHD0EIXrmbzVyD23HQjT79ApTMCVzLNlhMr/WpSJZ9tzzIGQ6Q39sAn MVIQ== MIME-Version: 1.0 X-Received: by 10.60.103.234 with SMTP id fz10mr58844456oeb.11.1426713652948; Wed, 18 Mar 2015 14:20:52 -0700 (PDT) Received: by 10.202.51.66 with HTTP; Wed, 18 Mar 2015 14:20:52 -0700 (PDT) In-Reply-To: <491C7497-BE3E-452B-A797-C7FC1102E7ED@gmail.com> References: <7081A75C-899A-4DB7-8D77-935A37B362D8@gmail.com> <5509957B.30600@pollere.com> <491C7497-BE3E-452B-A797-C7FC1102E7ED@gmail.com> Date: Wed, 18 Mar 2015 14:20:52 -0700 Message-ID: From: Dave Taht To: Jonathan Morton Content-Type: multipart/alternative; boundary=089e0115e8687af5e7051196a8f9 Cc: "codel@lists.bufferbloat.net" Subject: Re: [Codel] The next slice of cake X-BeenThere: codel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: CoDel AQM discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 18 Mar 2015 21:21:22 -0000 --089e0115e8687af5e7051196a8f9 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On Wed, Mar 18, 2015 at 8:20 AM, Jonathan Morton wrote: > > > On 18 Mar, 2015, at 17:10, Kathleen Nichols wrote= : > > > > How are you relating target delay to bandwidth? > > Essentially, I use 5ms as a minimum, and increase it if necessary to > accommodate a couple of MTU-sized packets at the shaping rate. This keep= s > things nicely under control at low bandwidths, and I find that cake remai= ns > useful and usable even at 64Kbps (without making even the usual adjustmen= ts > to host or link configuration for such low speeds). > In the cake2 (or maybe it was the unpublished cake3) version, I had a lighter weight version of the codel algorithm, that did not have a target parameter at all. Instead it just took the interval parameter and shifted it right 4 (yielding a target of 6.xms from an interval of 100ms) This saves on a memory access (and storage per queue!) , and I felt that any differences in behavior would be unnoticeable. And they were. This is also above the bound for cable-modem media access that greg white (rightly or wrongly) believed existed. So I have no problem in eliminating "target" entirely. Cake (without bandwidth shaping engaged) uses more cpu than fq_codel did and this was one of many optimizations I'd attempted (or successfully added). Cake with shaping is a bit less cpu than sqm-scripts htb + fq_codel + filters. It also looked like cake could be poured into gates, with a bit more research, and testing. > I can do this in cake because the shaping rate is known, whereas the pure > codel and fq_codel qdiscs do not have reliable link-speed information. As for this bit, we seemed to need to account for a MTU's worth of data at the lower speeds, and I did not explore what fiddling with the interval and auto-calc-ing the target did at these speeds, as yet. > - Jonathan Morton > > _______________________________________________ > Codel mailing list > Codel@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/codel > --=20 Dave T=C3=A4ht Let's make wifi fast, less jittery and reliable again! https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb --089e0115e8687af5e7051196a8f9 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable


On Wed, Mar 18, 2015 at 8:20 AM, Jonathan Morton <= chromatix99@gmai= l.com> wrote:

> On 18 Mar, 2015, at 17:10, Kathleen Nichols <nichols@pollere.com> wrote:
>
> How are you relating target delay to bandwidth?

Essentially, I use 5ms as a minimum, and increase it if necessary to= accommodate a couple of MTU-sized packets at the shaping rate.=C2=A0 This = keeps things nicely under control at low bandwidths, and I find that cake r= emains useful and usable even at 64Kbps (without making even the usual adju= stments to host or link configuration for such low speeds).

In the cake2 (or maybe it was the unpublished cake3) v= ersion, I had a lighter weight version of the codel algorithm, that did not= have a target parameter at all. Instead it just took the interval paramete= r and shifted it=C2=A0
right 4 (yielding a target of 6.xms from a= n interval of 100ms)

This saves on a memory access= (and storage per queue!) , and I felt that any differences in behavior wou= ld be unnoticeable. And they were. This is also above the bound for cable-m= odem media access that greg white (rightly or wrongly) believed existed. So= I have no problem in eliminating "target" entirely.
Cake (without bandwidth shaping engaged) uses more cpu than fq= _codel did and this was one of many optimizations I'd attempted (or suc= cessfully added). Cake with shaping is a bit less cpu than sqm-scripts htb = + fq_codel + filters.=C2=A0

It also looked like ca= ke could be poured into gates, with a bit more research, and testing.
=


I can do this in cake because the shaping rate is known, whereas the pure c= odel and fq_codel qdiscs do not have reliable link-speed information.

As for this bit, we seemed to need to account fo= r a MTU's worth of data at the lower speeds, and I did not explore what= fiddling with the interval and auto-calc-ing the target did at these speed= s, as yet.
=C2=A0
- Jonathan Morton

_______________________________________________
Codel mailing list
Codel@lists.bufferbloat.net<= /a>
= https://lists.bufferbloat.net/listinfo/codel



--
=
Dave T=C3=A4ht
Let's make wifi fast, = less jittery and reliable again!

https://plus= .google.com/u/0/107942175615993706558/posts/TVX3o84jjmb
--089e0115e8687af5e7051196a8f9--