From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp201.iad.emailsrvr.com (smtp201.iad.emailsrvr.com [207.97.245.201]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by huchra.bufferbloat.net (Postfix) with ESMTPS id 94BFD21F196 for ; Mon, 26 Nov 2012 07:04:13 -0800 (PST) Received: from localhost (localhost.localdomain [127.0.0.1]) by smtp30.relay.iad1a.emailsrvr.com (SMTP Server) with ESMTP id C1F00203FD; Mon, 26 Nov 2012 10:04:11 -0500 (EST) X-Virus-Scanned: OK Received: from legacy21.wa-web.iad1a (legacy21.wa-web.iad1a.rsapps.net [192.168.2.207]) by smtp30.relay.iad1a.emailsrvr.com (SMTP Server) with ESMTP id 9C44C202F0; Mon, 26 Nov 2012 10:04:11 -0500 (EST) Received: from reed.com (localhost.localdomain [127.0.0.1]) by legacy21.wa-web.iad1a (Postfix) with ESMTP id 8BDF71CE806A; Mon, 26 Nov 2012 10:04:11 -0500 (EST) Received: by apps.rackspace.com (Authenticated sender: dpreed@reed.com, from: dpreed@reed.com) with HTTP; Mon, 26 Nov 2012 10:04:11 -0500 (EST) Date: Mon, 26 Nov 2012 10:04:11 -0500 (EST) From: dpreed@reed.com To: "Michael Richardson" MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_20121126100411000000_54823" Importance: Normal X-Priority: 3 (Normal) X-Type: html In-Reply-To: <31933.1353939756@obiwan.sandelman.ca> References: <20121125232034.GF24680@merlins.org> <31933.1353939756@obiwan.sandelman.ca> Message-ID: <1353942251.571510886@apps.rackspace.com> X-Mailer: webmail7.0 Cc: cerowrt-users@lists.bufferbloat.net, cerowrt-devel@lists.bufferbloat.net Subject: Re: [Cerowrt-devel] [Cerowrt-users] QOS settings vs speedboost and random bandwidth 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: Mon, 26 Nov 2012 15:04:14 -0000 ------=_20121126100411000000_54823 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable =0AYou can use a small fraction of the capacity of the cable uplink path to= measure its queueing delay dynamically, and when it gets longer than laten= cy*"expected bitrate", reduce "expected bitrate".=0A =0AYou want to do this= *as quickly as possible*, so what you do is insert a "link monitor" task i= n the driver that sends tiny probe packets addressed to the nearest "loopba= ck point" you can find/create on the other side, and measure the RTT. You = can use, for example, the technique used by traceroute, which is to set the= hop count to the smallest number that causes a return ICMP packet to be se= nt, and send one of those periodically.=0A =0AI used this specific techniqu= e to cause my uplink queue to move back into my router, where I could manag= e it. You can also use it for the downlink queue measurement, but it doesn= 't move the queue into the router smoothly, instead you have to drop/ECN-ma= rk the IP frames coming in.=0A =0AThis can all be done between the IP layer= and layer 2. Since it exploits speedboost better, it might be worth addin= g as an option to cerowrt, so you don't have to set a speed limit explicitl= y when you have a single connection to the public Internet.=0A =0A-----Orig= inal Message-----=0AFrom: "Michael Richardson" =0ASent: M= onday, November 26, 2012 9:22am=0ATo: cerowrt-users@lists.bufferbloat.net, = cerowrt-devel@lists.bufferbloat.net=0ASubject: Re: [Cerowrt-devel] [Cerowrt= -users] QOS settings vs speedboost and random bandwidth=0A=0A=0A=0A>>>>> "D= ave" =3D=3D Dave Taht writes:=0A Dave> to move the fq= _codel algorithm into the cable modem, OR to have the=0A Dave> cable modem = exert flow control like ADSL modems do. Then have the=0A Dave> smarter devi= ce (the router) then do smarter queuing.=0A=0AI was unaware that DSL models= were doing this... is this with GbE=0Aethernet level XON/XOFF? Should I = assume that any DSL model with less=0Athan a GbE sync doesn't/can't have it= ? Any brands you'd like to=0Aacknowledge?=0A=0A-- =0A] He who is tir= ed of Weird Al is tired of life! | firewalls [=0A] Michael Ri= chardson, Sandelman Software Works, Ottawa, ON |net architect[=0A] mcr@s= andelman.ottawa.on.ca http://www.sandelman.ottawa.on.ca/ |device driver[=0A= Kyoto Plus: watch the video = =0A then sign the petition. =0A____________________________________________= ___=0ACerowrt-devel mailing list=0ACerowrt-devel@lists.bufferbloat.net=0Aht= tps://lists.bufferbloat.net/listinfo/cerowrt-devel ------=_20121126100411000000_54823 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

= You can use a small fraction of the capacity of the cable uplink path to me= asure its queueing delay dynamically, and when it gets longer than latency*= "expected bitrate", reduce "expected bitrate".

=0A

 

=0A

You want to do thi= s *as quickly as possible*, so what you do is insert a "link monitor" task = in the driver that sends tiny probe packets addressed to the nearest "loopb= ack point" you can find/create on the other side, and measure the RTT. = ; You can use, for example, the technique used by traceroute, which is to s= et the hop count to the smallest number that causes a return ICMP packet to= be sent, and send one of those periodically.

=0A

 

=0A

I used this specifi= c technique to cause my uplink queue to move back into my router, where I c= ould manage it.  You can also use it for the downlink queue measuremen= t, but it doesn't move the queue into the router smoothly, instead you have= to drop/ECN-mark the IP frames coming in.

=0A

 

=0A

This can all be done b= etween the IP layer and layer 2.  Since it exploits speedboost better,= it might be worth adding as an option to cerowrt, so you don't have to set= a speed limit explicitly when you have a single connection to the public I= nternet.

=0A

 

=0A

-----Original Message-----
From: "Michael Richardso= n" <mcr@sandelman.ca>
Sent: Monday, November 26, 2012 9:22am
To: cerowrt-users@lists.bufferbloat.net, cerowrt-devel@lists.bufferbloat.= net
Subject: Re: [Cerowrt-devel] [Cerowrt-users] QOS settings vs speed= boost and random bandwidth

=0A

>>>>> "Dave" =3D=3D Dave Taht <dave.taht@gmail.c= om> writes:
Dave> to move the fq_codel algorithm into the cable= modem, OR to have the
Dave> cable modem exert flow control like A= DSL modems do. Then have the
Dave> smarter device (the router) the= n do smarter queuing.

I was unaware that DSL models were doing t= his... is this with GbE
ethernet level XON/XOFF? Should I assume tha= t any DSL model with less
than a GbE sync doesn't/can't have it? Any = brands you'd like to
acknowledge?

--
] He who i= s tired of Weird Al is tired of life! | firewalls [
] Mi= chael Richardson, Sandelman Software Works, Ottawa, ON |net architect[] mcr@sandelman.ottawa.on.ca http://www.sandelman.ottawa.on.ca/ |device= driver[
Kyoto Plus: watch the video <http://www.youtube.com/watch= ?v=3Dkzx1ycLXQSE>
then sign the petition.
__________________= _____________________________
Cerowrt-devel mailing list
Cerowrt-= devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cer= owrt-devel
 ------=_20121126100411000000_54823--