From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-bk0-x230.google.com (mail-bk0-x230.google.com [IPv6:2a00:1450:4008:c01::230]) (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 B01EA21F1FC for ; Sun, 8 Dec 2013 11:01:27 -0800 (PST) Received: by mail-bk0-f48.google.com with SMTP id r7so156277bkg.21 for ; Sun, 08 Dec 2013 11:01:25 -0800 (PST) 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=wpMSsQMS9+dX8oyYK/aODYHB9vat5B5U2RBih0KlRSc=; b=lozU7xXl/lNsL92BPB9aOwiM/jIWnwl5w4vmGwAwUun9VNff3hQGZQJEfXdRiBaigi pl9WQi1iLvzIkrUBjijdkfQMaiBGji8EJCvzpyYnh+EFH842ZURYrmKYgzn2zbnJ0+Ze AE7KKKH1/QP2TzfbmFiu9ep9cNF5UNQ8gD40bInhOGrDgIO7pgPHNCK6z5o/TqxE1KlC cBpH72EfwHagWE8nldK9HHHTHIEqWN0F0YOAWin0My37mx6QgVeOA+JG1RglZRp4iMnB cQuWWB0MWNWN65K2lXdKcyx/1rqv1YDedJBZNgf92wS3SrJVGhBKeI6XxSyrq2TbCyz+ d8uA== MIME-Version: 1.0 X-Received: by 10.205.23.194 with SMTP id rb2mr1074371bkb.112.1386529285468; Sun, 08 Dec 2013 11:01:25 -0800 (PST) Received: by 10.205.18.135 with HTTP; Sun, 8 Dec 2013 11:01:25 -0800 (PST) Received: by 10.205.18.135 with HTTP; Sun, 8 Dec 2013 11:01:25 -0800 (PST) In-Reply-To: References: <20131203222559.GV8066@einstein.kenyonralph.com> <7ieh5pew2d.wl%jch@pps.univ-paris-diderot.fr> <87haakx1ev.wl%jch@pps.univ-paris-diderot.fr> <09D8F3A0-7172-4677-9887-119813E28740@gmx.de> <877gbfcw4t.wl%jch@pps.univ-paris-diderot.fr> Date: Sun, 8 Dec 2013 21:01:25 +0200 Message-ID: From: Jonathan Morton To: Sebastian Moeller Content-Type: multipart/alternative; boundary=20cf30363e4f8817f504ed0a81b3 Cc: bloat , Juliusz Chroboczek Subject: Re: [Bloat] curious..... 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: Sun, 08 Dec 2013 19:01:28 -0000 --20cf30363e4f8817f504ed0a81b3 Content-Type: text/plain; charset=ISO-8859-1 Data point: Annex M ADSL2 can be approximated as 10M down, 2M up in practice. Throw BitTorrent at that, and round-robin delay absolutely is relevant. ADSL1 connections will be even more so. Not everyone lives in a city in Scandinavia. So a simple tiered scheme which can distinguish VoIP from BitTorrent and both from general traffic, and applies fq_codel to each tier, is a good idea. - Jonathan Morton On 8 Dec 2013 20:49, "Sebastian Moeller" wrote: > Hi Juliusz, > > On Dec 8, 2013, at 14:25 , Juliusz Chroboczek < > jch@pps.univ-paris-diderot.fr> wrote: > > >>> The promise of fq_codel is that we can get rid of our prioritising > >>> hacks -- if we need that kind of features, then fq_codel has > >>> failed. > > > >> Is that really true? given enough concurrent flows, critical flows > >> might be delayed purely be the round robin scheduling of equally > >> "worthy" packets in fq_codel > > > > At 100 Mbit, one full-size Ethernet frame is 120us. This means that > > if you want your VoIP traffic to have less than 30ms delay, you should > > in principle reach your deadline as long as you have fewer than 250 > > congestion-limited flows at a given time. > > Well, is 250 enough and are the 250 really realistic in a > residential setting? Currently not doing much of anything my router has 142 > active connections (according to conntrack) so 250 might be on the low size > for a device that routes multiple devices. Plus, unfortunately, most > residential internet connections are asymmetric, so on the upload will > allow fewer congestion-limited concurrent flows before the round robin > delay will impede the VOIP session. (In Germany residential VDSL with > 100Mbit/s downlink will run at 40Mbit/s uplink, so hopefully not a big > issue, but most cable providers keep the upload below 10Mbit/s, typically > 5Mbit/s for 100Mbit/s download). So we talk about an order of magnitude > fewer flows required to make phone calls "interesting". > So I still think that for VoIP prioritizing might still be > required until supplied minimum bandwidth gets higher. > > > > >> so some residual priory system might still make sense... > > > > For throughput-sharing reasons, perhaps. For latency reasons, hopefully > not. > > Even at 1000 symmetric I still think it would be a good idea to > isolate really latency critical traffic from the rest, even if under normal > circumstances there should be no problem, I guess a "better safe than > sorry" approach. But, hey I do not do this for a living so I might be on > the wrong track here. > > best > Sebastian > > > > > -- Juliusz > > _______________________________________________ > Bloat mailing list > Bloat@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/bloat > --20cf30363e4f8817f504ed0a81b3 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable

Data point: Annex M ADSL2 can be approximated as 10M down, 2= M up in practice. Throw BitTorrent at that, and round-robin delay absolutel= y is relevant. ADSL1 connections will be even more so. Not everyone lives i= n a city in Scandinavia.

So a simple tiered scheme which can distinguish VoIP from Bi= tTorrent and both from general traffic, and applies fq_codel to each tier, = is a good idea.

- Jonathan Morton

On 8 Dec 2013 20:49, "Sebastian Moeller&quo= t; <moeller0@gmx.de> wrote:
Hi Juliusz,

On Dec 8, 2013, at 14:25 , Juliusz Chroboczek <jch@pps.univ-paris-diderot.fr> wrote:

>>> The promise of fq_codel is that we can get rid of our prioriti= sing
>>> hacks -- if we need that kind of features, then fq_codel has >>> failed.
>
>> Is that really true? given enough concurrent flows, critical flows=
>> might be delayed purely be the round robin scheduling of equally >> "worthy" packets in fq_codel
>
> At 100 Mbit, one full-size Ethernet frame is 120us. =A0This means that=
> if you want your VoIP traffic to have less than 30ms delay, you should=
> in principle reach your deadline as long as you have fewer than 250 > congestion-limited flows at a given time.

=A0 =A0 =A0 =A0 Well, is 250 enough and are the 250 really realistic in a r= esidential setting? Currently not doing much of anything my router has 142 = active connections (according to conntrack) so 250 might be on the low size= for a device that routes multiple devices. Plus, unfortunately, most resid= ential internet connections are asymmetric, so on the upload will allow few= er congestion-limited concurrent flows before the round robin delay will im= pede the VOIP session. (In Germany residential VDSL with 100Mbit/s downlink= will run at 40Mbit/s uplink, so hopefully not a big issue, but most cable = providers keep the upload below 10Mbit/s, typically 5Mbit/s for 100Mbit/s d= ownload). =A0So we talk about an order of magnitude fewer flows required to= make phone calls "interesting".
=A0 =A0 =A0 =A0 So I still think that for VoIP prioritizing might still be = required until supplied minimum bandwidth gets higher.

>
>> so some residual priory system might still make sense...
>
> For throughput-sharing reasons, perhaps. =A0For latency reasons, hopef= ully not.

=A0 =A0 =A0 =A0 Even at 1000 symmetric I still think it would be a good ide= a to isolate really latency critical traffic from the rest, even if under n= ormal circumstances there should be no problem, I guess a "better safe= than sorry" approach. But, hey I do not do this for a living so I mig= ht be on the wrong track here.

best
=A0 =A0 =A0 =A0 Sebastian

>
> -- Juliusz

_______________________________________________
Bloat mailing list
Bloat@lists.bufferbloat.net<= /a>
= https://lists.bufferbloat.net/listinfo/bloat
--20cf30363e4f8817f504ed0a81b3--