From detlef.bosau at web.de Mon Jul 5 05:27:21 2010 From: detlef.bosau at web.de (Detlef Bosau) Date: Mon, 05 Jul 2010 14:27:21 +0200 Subject: [e2e] CWND handling in idle periods Message-ID: <4C31CFA9.2070107@web.de> Hi, IIRC, TCP halves its CWND when being idle for a certain period of time. IIRC, there is an RFC where this is specified. Does someone happen do know the number? Thx Detlef -- ------------------------------------------------------------------------ Detlef Bosau Galileistra?e 30 70565 Stuttgart Tel.: +49 711 5208031 mobile: +49 172 6819937 skype: detlef.bosau ICQ: 566129673 detlef.bosau at web.de http://www.detlef-bosau.de ------------------------------------------------------------------------ From Wim.Lamotte at uhasselt.be Mon Jul 5 06:54:41 2010 From: Wim.Lamotte at uhasselt.be (Wim Lamotte) Date: Mon, 5 Jul 2010 15:54:41 +0200 Subject: [e2e] CWND handling in idle periods In-Reply-To: <4C31CFA9.2070107@web.de> References: <4C31CFA9.2070107@web.de> Message-ID: <002d01cb1c49$9dfb3ef0$d9f1bcd0$@Lamotte@uhasselt.be> Hi Detlef, Isn't this section 4.1 in RFC 5681? This section says that the cwnd should restart at the initial congestion window size. In RFC 2862, Handley et al. proposed "that after an idle period, the TCP sender should reduce its congestion window by half for every RTT that the flow has remained idle" - but this is an experimental extension afaik. Regards, Wim -- Prof. dr. Wim Lamotte Hasselt University Wetenschapspark 2 - 3590 Diepenbeek Belgium > -----Original Message----- > From: end2end-interest-bounces at postel.org [mailto:end2end-interest- > bounces at postel.org] On Behalf Of Detlef Bosau > Sent: maandag 5 juli 2010 14:27 > To: end2end-interest at postel.org > Subject: [e2e] CWND handling in idle periods > > Hi, > > IIRC, TCP halves its CWND when being idle for a certain period of time. > IIRC, there is an RFC where this is specified. Does someone happen do > know the number? > > Thx > > Detlef > > -- > ----------------------------------------------------------------------- > - > Detlef Bosau Galileistra?e 30 70565 > Stuttgart > Tel.: +49 711 5208031 mobile: +49 172 6819937 skype: > detlef.bosau > ICQ: 566129673 detlef.bosau at web.de http://www.detlef- > bosau.de > ----------------------------------------------------------------------- > - From johnwheffner at gmail.com Mon Jul 5 07:06:57 2010 From: johnwheffner at gmail.com (John Heffner) Date: Mon, 5 Jul 2010 10:06:57 -0400 Subject: [e2e] CWND handling in idle periods In-Reply-To: <4C31CFA9.2070107@web.de> References: <4C31CFA9.2070107@web.de> Message-ID: I think you're looking for RFC2861: TCP Congestion Window Validation. -John On Mon, Jul 5, 2010 at 8:27 AM, Detlef Bosau wrote: > Hi, > > IIRC, TCP halves its CWND when being idle for a certain period of time. > IIRC, there is an RFC where this is specified. Does someone happen do know > the number? > > Thx > > Detlef > > -- > ------------------------------------------------------------------------ > Detlef Bosau ? ? ? ? ? ? Galileistra?e 30 ? ? ? ? ? ? ? ?70565 Stuttgart > Tel.: +49 711 5208031 ? ?mobile: +49 172 6819937 ? ?skype: ?detlef.bosau > ICQ: 566129673 ? ? ? ? ? detlef.bosau at web.de ?http://www.detlef-bosau.de > ------------------------------------------------------------------------ > > > From arjuna at erg.abdn.ac.uk Mon Jul 5 23:38:29 2010 From: arjuna at erg.abdn.ac.uk (Arjuna) Date: Tue, 6 Jul 2010 07:38:29 +0100 Subject: [e2e] end2end-interest Digest, Vol 76, Issue 1 In-Reply-To: References: Message-ID: <000d01cb1cd5$dd2851d0$9778f570$@abdn.ac.uk> Dear Deltef, Its RFC 2861 - CWV Validation. FYI - we have been working on this for the last few years and have a PhD student who is working on this --bursty apps over TCP and how they should respond after idle/data-limited periods etc..Please do let us know if you require any info on this :) Regards Arjuna > -----Original Message----- > From: end2end-interest-bounces at postel.org [mailto:end2end-interest- > bounces at postel.org] On Behalf Of end2end-interest-request at postel.org > Sent: 05 July 2010 20:00 > To: end2end-interest at postel.org > Subject: end2end-interest Digest, Vol 76, Issue 1 > > Send end2end-interest mailing list submissions to > end2end-interest at postel.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://mailman.postel.org/mailman/listinfo/end2end-interest > or, via email, send a message with subject or body 'help' to > end2end-interest-request at postel.org > > You can reach the person managing the list at > end2end-interest-owner at postel.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of end2end-interest digest..." > > > Today's Topics: > > 1. CWND handling in idle periods (Detlef Bosau) > 2. Re: CWND handling in idle periods (Wim Lamotte) > 3. Re: CWND handling in idle periods (John Heffner) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Mon, 05 Jul 2010 14:27:21 +0200 > From: Detlef Bosau > Subject: [e2e] CWND handling in idle periods > To: end2end-interest at postel.org > Message-ID: <4C31CFA9.2070107 at web.de> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Hi, > > IIRC, TCP halves its CWND when being idle for a certain period of time. > IIRC, there is an RFC where this is specified. Does someone happen do > know the number? > > Thx > > Detlef > > -- > ----------------------------------------------------------------------- > - > Detlef Bosau Galileistra?e 30 70565 > Stuttgart > Tel.: +49 711 5208031 mobile: +49 172 6819937 skype: > detlef.bosau > ICQ: 566129673 detlef.bosau at web.de http://www.detlef- > bosau.de > ----------------------------------------------------------------------- > - > > > > > ------------------------------ > > Message: 2 > Date: Mon, 5 Jul 2010 15:54:41 +0200 > From: "Wim Lamotte" > Subject: Re: [e2e] CWND handling in idle periods > To: > Message-ID: <002d01cb1c49$9dfb3ef0$d9f1bcd0$@Lamotte at uhasselt.be> > Content-Type: text/plain; charset="iso-8859-1" > > Hi Detlef, > > Isn't this section 4.1 in RFC 5681? This section says that the cwnd > should > restart at the initial congestion window size. > > In RFC 2862, Handley et al. proposed "that after an idle period, the > TCP > sender should reduce its congestion window by half for every RTT that > the > flow has remained idle" - but this is an experimental extension afaik. > > Regards, > > Wim > -- > Prof. dr. Wim Lamotte > Hasselt University > Wetenschapspark 2 - 3590 Diepenbeek > Belgium > > > > -----Original Message----- > > From: end2end-interest-bounces at postel.org [mailto:end2end-interest- > > bounces at postel.org] On Behalf Of Detlef Bosau > > Sent: maandag 5 juli 2010 14:27 > > To: end2end-interest at postel.org > > Subject: [e2e] CWND handling in idle periods > > > > Hi, > > > > IIRC, TCP halves its CWND when being idle for a certain period of > time. > > IIRC, there is an RFC where this is specified. Does someone happen do > > know the number? > > > > Thx > > > > Detlef > > > > -- > > --------------------------------------------------------------------- > -- > > - > > Detlef Bosau Galileistra?e 30 70565 > > Stuttgart > > Tel.: +49 711 5208031 mobile: +49 172 6819937 skype: > > detlef.bosau > > ICQ: 566129673 detlef.bosau at web.de http://www.detlef- > > bosau.de > > --------------------------------------------------------------------- > -- > > - > > > > > > ------------------------------ > > Message: 3 > Date: Mon, 5 Jul 2010 10:06:57 -0400 > From: John Heffner > Subject: Re: [e2e] CWND handling in idle periods > To: Detlef Bosau > Cc: end2end-interest at postel.org > Message-ID: > > Content-Type: text/plain; charset=ISO-8859-1 > > I think you're looking for RFC2861: TCP Congestion Window Validation. > > -John > > > On Mon, Jul 5, 2010 at 8:27 AM, Detlef Bosau > wrote: > > Hi, > > > > IIRC, TCP halves its CWND when being idle for a certain period of > time. > > IIRC, there is an RFC where this is specified. Does someone happen do > know > > the number? > > > > Thx > > > > Detlef > > > > -- > > --------------------------------------------------------------------- > --- > > Detlef Bosau ? ? ? ? ? ? Galileistra?e 30 ? ? ? ? ? ? ? ?70565 > Stuttgart > > Tel.: +49 711 5208031 ? ?mobile: +49 172 6819937 ? ?skype: > ?detlef.bosau > > ICQ: 566129673 ? ? ? ? ? detlef.bosau at web.de ?http://www.detlef- > bosau.de > > --------------------------------------------------------------------- > --- > > > > > > > > > > ------------------------------ > > _______________________________________________ > end2end-interest mailing list > end2end-interest at postel.org > http://mailman.postel.org/mailman/listinfo/end2end-interest > > > End of end2end-interest Digest, Vol 76, Issue 1 > *********************************************** From sbmoon at kaist.edu Mon Jul 5 09:01:39 2010 From: sbmoon at kaist.edu (Sue Moon) Date: Tue, 6 Jul 2010 01:01:39 +0900 Subject: [e2e] ACM SIGCOMM 2010 Call for Participation Message-ID: <000b01cb1c5b$5c9a7860$15cf6920$@edu> *** Our apologies if you receive multiple copies of this message. *** ------------------------------------------------------------- ACM SIGCOMM 2010: Call For Participation http://conferences.sigcomm.org/sigcomm/2010/index.php ------------------------------------------------------------- Welcome The 2010 ACM SIGCOMM organization and program committees are delighted to invite you to participate in the first ever SIGCOMM in India. SIGCOMM 2010 is the annual conference of the ACM Special Interest Group on Data Communication (SIGCOMM) on the applications, technologies, architectures, and protocols for computer communication. This year we have an exciting technical program of 33 papers, 5 workshops, 22 posters and 20 demos. We will have a rich social program with a reception and two dinners (including the banquet), as well as a fun outrageous opinion session. Thanks to our corporate, government and other supporters we have strived to keep the conference accessible to everyone with a low registration fee structure. There are also a variety of sightseeing opportunities attendees can take advantage of before or after the conference. Please note the early registration and hotel cut-off dates below. Register TODAY! Important Dates Early registration deadline: July 25th, 2010 - http://conferences.sigcomm.org/sigcomm/2010/registration.php Early hotel booking cut-off dates for room upgrades at the main venue: - Le M?ridien Hotel: http://conferences.sigcomm.org/sigcomm/2010/hotel.php - July 10th for Royal Club Room (two-way transfers to the airport included) - July 31st for Superior Room (one-way pick-up only) - After July 31st, Executive Room (no pick-up) Some other hotels' reservation deadlines: July 20th to 31, 2010 - Alternate hotels (including a very affordable budget option): http://conferences.sigcomm.org/sigcomm/2010/althotel.php All non-Indian citizens (except for nationals of Nepal and Bhutan) require a visa to enter India. Allow yourself sufficient time for visa processing. For further information, visit http://conferences.sigcomm.org/sigcomm/2010/visainfo.php Shivkumar Kalyanaraman and Venkat Padmanabhan SIGCOMM 2010 General Co-Chairs From ycheng at google.com Tue Jul 20 11:00:24 2010 From: ycheng at google.com (Yuchung Cheng) Date: Tue, 20 Jul 2010 11:00:24 -0700 Subject: [e2e] [78attendees] wanted: your old NAT home router In-Reply-To: References: <03C884AE-530A-4C24-B686-25508D879619@nokia.com> Message-ID: Hi Lars, This is very interesting study! it provides many useful information for the TCP research at Google. The queuing and processing delays on slide 12 (or figure 9 in the paper) are mostly <50ms, but the following paper "Characterizing residential broadband networks" published in IMC 2007 measures much higher queuing length ranging from 100ms to a few secs ( http://portal.acm.org/citation.cfm?id=1298306.1298313 figure 13). Any idea about the differences? Thanks, Yuchung On Fri, Jul 9, 2010 at 12:15 AM, Lars Eggert wrote: > Hi, > > a quick status update. We now have received over 100 donated home gateways, > plus a DSLAM. The students are on their summer break, after which we'll > start running a significantly expanded set of tests over this much larger > population of devices. > > Many of yo have donated boxes and suggested more experiments and better > ways of performing our current tests - thank you! > > In case you are attending IETF-78 in Maastricht and would like to donate a > home gateway, simply bring it. (Or contact me now for shipping details; no > cost to you.) > > We're especially interested in devices from outside the EU and North > America, or any other model we may not have yet (see > http://fit.nokia.com/lars/tmp/2010-hgw-study-devices.txt). And we're still > lacking a CMTS for testing cable modems... > > See you in Maastricht, > Lars > > On 2010-6-2, at 18:36, Lars Eggert wrote: > > Hi, > > > > FYI, a first report with test results for 34 devices is available at > http://fit.nokia.com/lars/tmp/2010-hgw-study.pdf. Slides that summarize > the results are at http://fit.nokia.com/lars/tmp/2010-hgw-study-slides.pdf > . > > > > We have received another 30-odd devices as donations, which we'll add to > the testbed and include in a follow-up study. > > > > If you have an unused, spare home gateway to donate to this effort, > please contact us at nat-study at fit.nokia.com. We're also interested in > obtaining a DSLAM and a CMTS. > > > > Thanks, > > Lars > > > > On 2010-4-29, at 12:34, Lars Eggert wrote: > >> Hi, > >> > >> for a measurement study done together with Markku Kojo's team at the > University of Helsinki, we're looking to collect as many different NAT home > routers as possible. If you have an old clunker lying around somewhere, > please contact me off-list. I'll cover shipping via DHL. Feel free to > forward this email as you see fit. > >> > >> The boxes will find a permanent home at the University of Helsinki. > Study results will be published openly. The intent is that this collection > become a resource for the community to be shared for future studies. > >> > >> Caveat: The boxes should NAT between Ethernet interfaces - we don't have > DSL or cable access equipment in the lab setup at the moment. > >> > >> Thanks, > >> Lars > > > > > _______________________________________________ > 78attendees mailing list > 78attendees at ietf.org > https://www.ietf.org/mailman/listinfo/78attendees > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.postel.org/pipermail/end2end-interest/attachments/20100720/5d36bcf6/attachment.html