[ NNSquad ] Re: [IP] BitTorrent uTorrent 2.0 uTP will self-throttle to protect networks

[travis kalanick <travisk@gmail.com>]

All,

 

Individual use cases are a good start to determining if the claims are true, but they are anecdotal at best.  The bottom line is that the only way to really determine whether any of their claims are true is to do broad-based gathering of empirical data of network utilization. 

 

As some of you may know, I was founder and CEO of a company called Red Swoosh, a P2P-CDN that Akamai acquired a couple of years ago.  The difficulty I have with this discussion is the lack of information that we have on what exactly Bittorrent's methods for internal network throttling are in their new client. 

 

I can tell you from experience that determining self-induced congestion on a network by the client that is supposedly creating that congestion is incredibly difficult from a technical perspective and is rife with client throttling policy conundrums.  How do you define "too much congestion", who's fault is it when that congestion occurs, and what are the policies you want to put into place to throttle back that congestion?  Those are the tough policy decisions.  Next up are the very challenging technical implementation problems.  Do you start creating open-sockets on the network to determine local line usage and sharing?  How do you get around security hurdles and issues to do so?  Do you do regular "out of band" testing of the network to verify what the network can handle at different times of the day? How does the user feel about you using his bandwidth to do serious download testing?  How do you do the testing in a way that doesn't affect what the user is doing on his machine? Maybe you use uPnP parameters to determine what's going on at the gateway, but proper support of that is scant at best.  You could keep it simple and just use absolute measurments of ping times, packet loss, etc., but why should an application arbitrarily determine it is a second-class network citizen, subjugating itself to other bittorrent applications that do not have the throttling?  Certainly any throttling will push users to either disable the uTorrent throttling if that's possible or maybe users will simply move over to another bittorrent application.

 

It's good to be the nice network guy, but I think the devil is in the details here.  Without those details and/or broad-based empirical evidence, I usually chalk it up as vapor-hype.

 

Thanks,

 

Travis

 

 

On Mon, Nov 2, 2009 at 2:33 PM, Vint Cerf <vint@google.com> wrote:

George,

This discussion suggests that users should have something to say about the priority of packet flows WITHIN the capacity they are paying for (capital letters just in lieu of italics; I am not shouting). If the access ISP can do traffic shaping to keep users within their pro-rata envelopes and also respond to user-specified priority, I would think we would be moving toward a balance that seems useful.

vint

On Nov 2, 2009, at 1:54 PM, George Ou wrote:

I’ve published my results here.

http://www.digitalsociety.org/2009/11/analysis-of-bittorrent-utp-congestion-avoidance/

 

Dr. Reed.  Your use of the words “rhetoric” and “tricks” aren’t very useful to this discussion, and I would take issue with your comments.

 

1.       BitTorrent still hogs over 90% of my broadband connection over HTTP.  This has significant ramifications beyond just real-time applications like VoIP and online gaming.

2.      You shouldn’t be so quick to discount VoIP and online gamers.  A very large number of BitTorrent (or any P2P app) users also do online gaming and VoIP, and they’re forced to shut down their P2P application when the use VoIP or game and that actually hurts the P2P upload and download throughput for the entire P2P community since there are fewer seeders.

3.      Don’t conflate wireless with wired broadband.  Just because 150 ms ping for wireless is best case doesn’t make 70 ms additional on a wired network bearable for online gaming.  Maybe you’re different, but I don’t know any gamer that will put up with an additional 70 ms if they can help it.  I thought it would be tolerable for VoIP, but my Lingo VoIP phone service drops a significant amount of audio even when I merely upload with BitTorrent.

 

 

 

George Ou

 

From: nnsquad-bounces+george_ou=lanarchitect.net@nnsquad.org [mailto:nnsquad-bounces+george_ou=lanarchitect.net@nnsquad.org] On Behalf OfDavid P. Reed

Sent: Monday, November 02, 2009 8:43 AM
To: 'NNSquad'
Cc: Lauren Weinstein
Subject: [ NNSquad ] Re: [IP] BitTorrent uTorrent 2.0 uTP will self-throttle to protect networks

 

I find the word games/rhetorical tricks that Ou and Bennett use fascinating.  We'll see whether Farber posts my response below.

George Ou wrote:

Subject: RE: [ NNSquad ] BitTorrent uTorrent 2.0 uTP will self-throttle to protect    networks 

Too bad nobody ever bothered to test if these claims actually hold water 
before repeating them endlessly.  I just tested uTorrent version 2 build 
16850 today and it still grabs all the bandwidth and jacks up the ping to 
unbearable levels for online gaming and VoIP.  It certainly does NOT protect 
my network.

I will do some testing myself, because I am curious about the mechanism in uTorrent 2.0.   I do note that "unbearable levels for online gaming and VoIP" is an interesting statement.

If true that means that ping times might be 100 msec or more.   Now, since I have been recently measuring ping times on networks where there are no "uTorrent" or other P2P services running, I can tell you that on a variety of commercial providers, 150 msec. ping times are common - and on ATT 3G in several cities, there are stable ping times that can be measured that are on the order of 2000-5000 msec.

So the "data" presented by Mr. Ou represents a very, very interesting choice of phrase.   Say that it is "unbearable" for two of the most sensitive-to-latency applications (only).   

I would, myself, stick to scientific measurements: how many milliseconds?   Clearly he has measured that data.   But I presume the hope of a talented columnist is to get the word "unbearable" to stick in the mind, and leave the "bumper sticker" impression without the qualifying information.

Rhetorical trickery?  You be the judge.   I'm gonna report numbers.