Hi! Sorry for the off topic (and the stupidity of the question), but could you briefly explain me what's wrong with the use of UDP?
Thanks! On Sun, Nov 28, 2010 at 12:00 AM, <p2p-hackers-requ...@lists.zooko.com>wrote: > Send p2p-hackers mailing list submissions to > p2p-hackers@lists.zooko.com > > To subscribe or unsubscribe via the World Wide Web, visit > http://lists.zooko.com/mailman/listinfo/p2p-hackers > or, via email, send a message with subject or body 'help' to > p2p-hackers-requ...@lists.zooko.com > > You can reach the person managing the list at > p2p-hackers-ow...@lists.zooko.com > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of p2p-hackers digest..." > > > Today's Topics: > > 1. Google cheating on TCP slow start (David Barrett) > 2. Re: Google cheating on TCP slow start (Ray Dillinger) > 3. Re: Google cheating on TCP slow start (Michael Blizek) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Fri, 26 Nov 2010 15:59:14 -0700 > From: David Barrett <dbarr...@quinthar.com> > Subject: [p2p-hackers] Google cheating on TCP slow start > To: p2p-hackers@lists.zooko.com > Message-ID: <4cf03bc2.7000...@quinthar.com> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Interesting article: > > http://blog.benstrong.com/2010/11/google-and-microsoft-cheat-on-slow.html > > I know a lot of people on this list are interested in this topic. But > I'm curious: if all sites were to start adopting *ahem* "alternative" > congestion strategies like this, would would the real-world > ramifications be? Indeed, it seems reasonable to assume that before > long it'll be a standard Apache option to do what Google does. > > Is this the end of the gentleman's internet? Should ISPs detect and > block/throttle this behavior -- essentially punishing (or overriding) > this type of behavior to re-establish normalcy? > > -david > > > ------------------------------ > > Message: 2 > Date: Fri, 26 Nov 2010 17:42:13 -0800 > From: Ray Dillinger <b...@sonic.net> > Subject: Re: [p2p-hackers] Google cheating on TCP slow start > To: theory and practice of decentralized computer networks > <p2p-hackers@lists.zooko.com> > Message-ID: <1290822133.28318.10.ca...@janus.pagansexcult.org> > Content-Type: text/plain > > On Fri, 2010-11-26 at 15:59 -0700, David Barrett wrote: > > Interesting article: > > > > > http://blog.benstrong.com/2010/11/google-and-microsoft-cheat-on-slow.html > > > > I know a lot of people on this list are interested in this topic. But > > I'm curious: if all sites were to start adopting *ahem* "alternative" > > congestion strategies like this, would would the real-world > > ramifications be? Indeed, it seems reasonable to assume that before > > long it'll be a standard Apache option to do what Google does. > > > > Is this the end of the gentleman's internet? Should ISPs detect and > > block/throttle this behavior -- essentially punishing (or overriding) > > this type of behavior to re-establish normalcy? > > > > -david > > > FWIW, I'm in favor of anything that reduces the number of roundtrips. > While I see the rationale behind the slow-start algorithm in HTTP, the > initial window size (and packet size) currently in use is ridiculous. > > Google is right that the IW size should be increased to (at least) 10 > packets. I think I'd have gone for 32 actually. Someone should write > an RFC to change it. > > This reflects the current reality that network latency has become by > far the limiting factor on speed; it's not (or it's rarely) the size > of data being transmitted that limits speed anymore; it's mainly the > number of roundtrip delays. > > Skipping the slow-start algorithm all together as MS is doing is > dubious at best, though. > > Bear > > > > > > > ------------------------------ > > Message: 3 > Date: Sat, 27 Nov 2010 08:04:49 +0100 > From: Michael Blizek <mic...@michaelblizek.twilightparadox.com> > Subject: Re: [p2p-hackers] Google cheating on TCP slow start > To: David Barrett <dbarr...@quinthar.com> > Cc: p2p-hackers@lists.zooko.com > Message-ID: <20101127070449.ga10...@michaelblizek.twilightparadox.com> > Content-Type: text/plain; charset=us-ascii > > Hi! > > On 15:59 Fri 26 Nov , David Barrett wrote: > > Interesting article: > > > > > http://blog.benstrong.com/2010/11/google-and-microsoft-cheat-on-slow.html > > > > I know a lot of people on this list are interested in this topic. But > > I'm curious: if all sites were to start adopting *ahem* "alternative" > > congestion strategies like this, would would the real-world > > ramifications be? Indeed, it seems reasonable to assume that before > > long it'll be a standard Apache option to do what Google does. > > The congestion avoidance is broken to begin with. There are not even "deep" > hacks like this needed to attack fairness. What Google does is actually not > that bad either. They do not even increase their "fair share" of bandwidth. > They could/might do this as well, but the article does not give any hint of > this. They just speed up connection "setup". Many users create much more > harm > with download managers to download big files over several connections at > once. > All of this just shows how badly congestion avoidance scales with so called > long fat pipes (high throughput and high latency). > > > Is this the end of the gentleman's internet? > > No, I do not think so. You need a backbone which fast enough avoid > congestion. > If your backbone is too slow, you either make it faster or limit the user > bandwidth. Relying to congestion avoidance to create fairness will not > work. > If you want to do this, replace IP with something which does bandwidth > "allocation" on the routers and not on the edges. > > > Should ISPs detect and > > block/throttle this behavior -- essentially punishing (or overriding) > > this type of behavior to re-establish normalcy? > > This is very hard. The fairness problem is not just about congestion > avoidance > hacks. You would also have to "punish" all applications which have many > open > connections or use UDP. This includes not only these download managers, but > also most P2P and server applications, VoIP (UDP streaming), multiplayer > games, web browsers and almost any other application which does anything on > the internet. Moreover without these "abuses" high internet bandwidths will > not even be possible. I would rather say that it is even within the > *interest* > of ISPs that people do such hacks, because otherwise they could not even > sell > the bandwidths they do today. > > -Michi > -- > programing a layer 3+4 network protocol for mesh networks > see http://michaelblizek.twilightparadox.com > > > > ------------------------------ > > _______________________________________________ > p2p-hackers mailing list > p2p-hackers@lists.zooko.com > http://lists.zooko.com/mailman/listinfo/p2p-hackers > > > End of p2p-hackers Digest, Vol 51, Issue 8 > ****************************************** > -- Jérôme Prudent
_______________________________________________ p2p-hackers mailing list p2p-hackers@lists.zooko.com http://lists.zooko.com/mailman/listinfo/p2p-hackers
