From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qa0-x22c.google.com (mail-qa0-x22c.google.com [IPv6:2607:f8b0:400d:c00::22c]) (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 2B3B920111B for ; Thu, 5 Dec 2013 12:45:30 -0800 (PST) Received: by mail-qa0-f44.google.com with SMTP id i13so186017qae.10 for ; Thu, 05 Dec 2013 12:45:29 -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=81pjoD1WEypnYim0q3hAronh28lgB1eAqThLxfvOU9w=; b=Ze7sYaybeMr6EL62d0S4YU3nzTUkRm/txpIYfyG3bUqnnNTOdRKcTmMugpb4w3rKQi 6B80uOwusAMo4FR9rMzWgjJpH8cCqcXSA7hXpfhzUOyPphPINfUSJGnT4smWyjkEH1Gh 2IoFm61tKRdZdw4UtaEKhzGDGjpyS+XbzNqR9L4XAGKkmlBHlVRlKekHt+X+VMQUNVj/ sTGkU/le8pB6Yrz0fqYwHgXtV3n3l7Zvur9rqGUNJVoo29t8uMbGVeAUoHzuhEVTrhIP sBekwehwbCPcq6AXd9qLpKOQRgFtqIcTik0NNM+7U7GscHsh/47TJrMqL5Xw/sv0/6J8 RPbA== MIME-Version: 1.0 X-Received: by 10.224.12.134 with SMTP id x6mr40482911qax.57.1386276329119; Thu, 05 Dec 2013 12:45:29 -0800 (PST) Received: by 10.96.62.66 with HTTP; Thu, 5 Dec 2013 12:45:29 -0800 (PST) In-Reply-To: References: Date: Thu, 5 Dec 2013 14:45:29 -0600 Message-ID: From: Dong Mo To: Anirudh Sivaraman Content-Type: multipart/alternative; boundary=089e013a07e428836504eccf9c25 Cc: codel@lists.bufferbloat.net Subject: Re: [Codel] How to make a bufferbloat fairqueue with tc qdisc 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: Thu, 05 Dec 2013 20:45:30 -0000 --089e013a07e428836504eccf9c25 Content-Type: text/plain; charset=ISO-8859-1 Thanks. I made a wrong observation for the last email that I used ping to measure the latency of TCP and was expecting to see a latency explode. So do you know a way to measure the running TCP's packet latency experienced at the qdisc? Thanks a lot -Mo 2013/12/5 Anirudh Sivaraman > On Thu, Dec 5, 2013 at 2:52 PM, Dong Mo wrote: > > Dear List, > > I am trying to make a bufferbloat fairqueue with qdisc and compare that > with > > fq_codel. > > However, I am now having some hard time doing it. > > > > The commands I am using are: > > > > tc qdisc add dev $IFACE root handle 1: htb default 1 > > tc class add dev $IFACE parent 1: classid 1:1 htb rate 40mbit > > tc qdisc del dev eth4 parent 1:1 handle 10: sfq perturb 10 > > > > And sending packet from the sender, where sfq qdisc is set, to the > receiver, > > the queuing delay is quite small and is approximately the same as when > > fq_codel enabled. > > > > However, if I am not using sfq (only htb set), and send only one flow > over > > this link, the delay will explode again. > > > > So is sfq here cut off the queue length somehow? How should I make a > > bufferbloat fair queue on linux box using tc qdisc? > > > > You could start with the fq_codel qdisc and set the codel interval and > target parameters to absurdly large values (say 1000 seconds each) to > simulate a qdisc that has fair queuing but almost never drops a > packet. > > Anirudh > > > Thanks > > > > -Mo Dong > > > > _______________________________________________ > > Codel mailing list > > Codel@lists.bufferbloat.net > > https://lists.bufferbloat.net/listinfo/codel > > > --089e013a07e428836504eccf9c25 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Thanks. I made a wrong observation for the last email that= I used ping to measure the latency of TCP and was expecting to see a laten= cy explode.

So do you know a way to measure the running = TCP's packet latency experienced at the qdisc?
Thanks a lot
-Mo




2013/12/5 Aniru= dh Sivaraman <sk.anirudh@gmail.com>
On T= hu, Dec 5, 2013 at 2:52 PM, Dong Mo <montedong@gmail.com> wrote:
> Dear List,
> I am trying to make a bufferbloat fairqueue with qdisc and compare tha= t with
> fq_codel.
> However, I am now having some hard time doing it.
>
> The commands I am using are:
>
> tc qdisc add dev $IFACE root handle 1: htb default 1
> tc class add dev $IFACE parent 1: classid 1:1 htb rate 40mbit
> tc qdisc del dev eth4 parent 1:1 handle 10: sfq perturb 10
>
> And sending packet from the sender, where sfq qdisc is set, to the rec= eiver,
> the queuing delay is quite small and is approximately the same as when=
> fq_codel enabled.
>
> However, if I am not using sfq (only htb set), and send only one flow = over
> this link, the delay will explode again.
>
> So is sfq here cut off the queue length somehow? How should I make a > bufferbloat fair queue on linux box using tc qdisc?
>

You could start with the fq_codel qdisc and set the codel inter= val and
target parameters to absurdly large values (say 1000 seconds each) to
simulate a qdisc that has fair queuing but almost never drops a
packet.

Anirudh

> Thanks
>
> -Mo Dong
>
> _______________________________________________
> Codel mailing list
> Codel@lists.bufferbloat= .net
> https://lists.bufferbloat.net/listinfo/codel
>

--089e013a07e428836504eccf9c25--