From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp1.unina.it (smtp1.unina.it [192.132.34.61]) (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 A496220032D for ; Fri, 8 Jul 2011 02:22:04 -0700 (PDT) Received: from mail-qy0-f178.google.com (mail-qy0-f178.google.com [209.85.216.178]) (authenticated bits=0) by smtp1.unina.it (8.14.4/8.14.4) with ESMTP id p689uVIU031900 (version=TLSv1/SSLv3 cipher=RC4-SHA bits=128 verify=OK) for ; Fri, 8 Jul 2011 11:56:32 +0200 Received: by qyk27 with SMTP id 27so1242486qyk.16 for ; Fri, 08 Jul 2011 02:56:30 -0700 (PDT) MIME-Version: 1.0 Received: by 10.229.241.205 with SMTP id lf13mr1361334qcb.292.1310118990937; Fri, 08 Jul 2011 02:56:30 -0700 (PDT) Received: by 10.229.100.72 with HTTP; Fri, 8 Jul 2011 02:56:30 -0700 (PDT) In-Reply-To: References: Date: Fri, 8 Jul 2011 11:56:30 +0200 Message-ID: From: Walter de Donato To: Dave Taht Content-Type: multipart/alternative; boundary=0016363b90d2117ffe04a78bd902 X-Mailman-Approved-At: Sat, 09 Jul 2011 08:05:37 -0700 Cc: bismark-devel@lists.bufferbloat.net Subject: Re: [Bismark-devel] modelling powerboost X-BeenThere: bismark-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: BISMark related software development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 08 Jul 2011 09:22:05 -0000 --0016363b90d2117ffe04a78bd902 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable We made some analysis about Powerboost in our SIGCOMM paper and it looks like a couple of TBFs with different bucket sizes (then not just one shaping rate). >From our results ~5MB is the bucket size for upstream and ~10MB for downstream. It shouldn't be too difficult to setup a such htb. Probably, to better model its real behavior. the analysis should be extende= d to a wider set of users. The related patent (US 7,289,447 B2) clearly says that its behavior can be dynamically changed depending on the network conditions. -Walter 2011/7/7 Dave Taht > It would be good to be able to have a working htb model of powerboost. > > ---------- Forwarded message ---------- > From: elided > Date: Thu, Jul 7, 2011 at 1:10 PM > Subject: Comcast Powerboost parameters > To: Dave Taht , Jim Gettys > > > Well, I poked around a bit and found the information was a little > harder to find than I remembered. > Here are three data points, none of them seem to be completely definitive= : > > http://customer.comcast.com/Pages/FAQListViewer.aspx?topic=3DInternet&fol= der=3D8b2fc392-4cde-4750-ba34-051cd5feacf0 > =85hopefully that URL will work for you. Click on How long does the > PowerBoost burst last. It indicates 5MB upstream token bucket burst. > http://www.dslreports.com/faq/14520 > =85Indicates 10MB upstream token bucket burst > http://www.cc.gatech.edu/~partha/shaperprobe-TR.pdf > =85table 2 indicates 5 or 10 MB depending on service tier. > > Seems as though it is likely either 5MB or 10MB. A fairly simple test > could validate which one your connection has. > -Greg > > > > -- > Dave T=E4ht > SKYPE: davetaht > US Tel: 1-239-829-5608 > http://the-edge.blogspot.com > _______________________________________________ > Bismark-devel mailing list > Bismark-devel@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/bismark-devel > _______________________________________________ > bismark-devel mailing list > bismark-devel@projectbismark.net > http://lists.noise.gatech.edu/listinfo/bismark-devel > > --=20 Walter de Donato, PhD Student Dipartimento di Informatica e Sistemistica Universit=E0 degli Studi di Napoli "Federico II" Via Claudio 21 -- 80125 Napoli (Italy) Phone: +39 081 76 83821 - Fax: +39 081 76 83816 Email: walter.dedonato@unina.it WWW: http://wpage.unina.it/walter.dedonato --0016363b90d2117ffe04a78bd902 Content-Type: text/html; charset=windows-1252 Content-Transfer-Encoding: quoted-printable We made some analysis about Powerboost in our SIGCOMM paper
and it look= s like a couple of TBFs with different bucket sizes (then not just one shap= ing rate).
From our results ~5MB is the bucket size for upstream = and ~10MB for downstream.

It shouldn't be too difficult to setup a such htb.<= /div>
Probably, to better model its real behavior. the analysis should = be extended to a wider set of users.

The related p= atent (US 7,289,447 B2) clearly says that its behavior can be=A0
dynamically changed=A0depending on the network conditions.
<= br>
-Walter

2011/7/7 Dave Taht = <dave.taht@gmai= l.com>
It would be good to be able to have a worki= ng htb model of powerboost.

---------- Forwarded message ----------
From: elided
Date: Thu, Jul 7, 2011 at 1:10 PM
Subject: Comcast Powerboost parameters
To: Dave Taht <dave.taht@gmail.co= m>, Jim Gettys <jg@freedesk= top.org>


Well, I poked around a bit and found the information was a little
harder to find than I remembered.
Here are three data points, none of them seem to be completely definitive:<= br> = http://customer.comcast.com/Pages/FAQListViewer.aspx?topic=3DInternet&f= older=3D8b2fc392-4cde-4750-ba34-051cd5feacf0
=85hopefully that URL will work for you. =A0Click on How long does the
PowerBoost burst last. =A0It indicates 5MB upstream token bucket burst.
http://ww= w.dslreports.com/faq/14520
=85Indicates 10MB upstream token bucket burst
http://www.cc.gatech.edu/~partha/shaperprobe-TR.pdf
=85table 2 indicates 5 or 10 MB depending on service tier.

Seems as though it is likely=A0either 5MB or 10MB. =A0A fairly simple test<= br> could validate which one your connection has.
-Greg



--
Dave T=E4ht
SKYPE: davetaht
US Tel: 1-239-829-5608
http://the-edge.= blogspot.com
_______________________________________________
Bismark-devel mailing list
Bismark-devel@lists.= bufferbloat.net
https://lists.bufferbloat.net/listinfo/bismark-devel
_______________________________________________
bismark-devel mailing list
bismark-devel@projectbi= smark.net
http://lists.noise.gatech.edu/listinfo/bismark-devel




--
Walter de Donato, PhD S= tudent
Dipartimento di Informatica e Sistemistica
Universit=E0=A0 deg= li Studi di Napoli "Federico II"
Via Claudio 21 -- 80125 Napol= i (Italy)
Phone: +39 081 76 83821 - Fax: +39 081 76 83816
Email: walter.dedonato@unina.itWWW: = http://wpage.unina.it/walter.dedonato

--0016363b90d2117ffe04a78bd902--