site Search:


 
   
story category
Using The New UDP BitTorrent To Dodge ISP Throttling
Looking more closely at works, and what won't...
by Karl Bode Friday 05-Dec-2008 tags: Fileswapping · business · net-neutrality
We recently noted that uTorrent, now owned by BitTorrent, released a new version of their client that lays uTP, the micro transport protocol, on top of UDP. That decision results in better congestion control, but it also prevents the kind of TCP RST packet attacks Comcast has used to throttle upstream P2P traffic, and in some cases allows users to skirt ISP throttling. BitTorrent insists to us their intention was a more Internet and network-friendly client, not to evade throttling.

While it may not have been BitTorrent's intent, our resident throttled Bell Canada & TekSavvy customers are using the client to dodge ISP throttling, albeit with mixed results. Many users who only saw P2P download speeds of between 20-50k/s, are now seeing speeds closer to 150-300k/s. Others however, claim the new beta client, which BitTorrent reiterates isn't yet prepped for general release, makes no difference in reducing throttling.

To be clear, bypassing Bell Canada throttling is not anywhere in the alpha's design motivation. It is just a coincidence.
-BitTorrent's Simon Morris
According to network expert Robb Topolski, the BBR user who first discovered Comcast's use of forged packets to kill upstream P2P delivery, the new client should help all users skirt ISPs who are using RST packet resets to cripple P2P (he's written a guide in our forums, here).

However, it won't be of much use to those hoping to evade Comcast's new protocol agnostic throttling system, which throttles a user back if a particular CMTS port is congested, and if that user has been consuming more than 70% of his allotted throughput for more than fifteen minutes.

Regardless, Topolski says the new client is designed to avoid congestion. "Users using uTP will not likely ever trigger the 70% threshold on a congested network port," he says. "uTP will detect the congestion condition through increased round-trip-times and start backing off before any packets get de-prioritized." He notes that while Comcast's throttling system takes 15 minutes to kick in, uTP does it in real time.

By this point you're probably grokking the larger point that the new client is actually fairly friendly to congested networks and the Internet, contrary to the hyperbolic tirade by Richard Bennett that garnered the Register oodles of hits last week as outlet after outlet pointed out that Bennett was wrong. Bennett's primary goal was to use the uTorrent announcement as a lectern for his anti-network neutrality positions, suggesting that neutrality laws would block ISPs from filtering any shift to a UDP-driven P2P platform -- causing the Internet to explode.

Bennett contacted me this week to suggest that the title of our original post on the new uTorrent client (New UDP uTorrent Takes Aim At Throttling), was at least partially responsible for his proclamation that Bittorrent had "declared war on VoIP, gamers," and was ushering forth digital oblivion. Bennett has since penned a follow up piece at El Reg that takes a drastically different tone from the first (the comments are worth reading).

I went back to the source to confirm that evading throttling was nowhere on BitTorrent's radar when designing the new uTorrent client, and spoke to Simon Morris, head of Product Management at BitTorrent.

"To be clear, bypassing Bell Canada throttling is not anywhere in the alpha's design motivation," says Morris, head of Product Management at BitTorrent. "It is just a coincidence." According to Morris, the client really won't be all that useful for anybody not dealing with throttling more sophisticated that TCP RST resets.

"If an ISP were relying completely on forged TCP reset packets to throttle Bittorrent traffic, then uTP would certainly evade this method," says Morris. But he notes there's some serious limitations to this. "Our client only talks uTP to other clients that talk uTP," he says - noting that the forged-TCP-reset-packet throttling method "would only be evaded completely in swarms consisting entirely of our new uTorrent 1.9 alpha client and uTorrent 1.8 clients" (which accept incoming uTP connections but default to TCP when trying to make outgoing connections).

"Of course there are many Bittorrent clients out there, and for now only some uTorrent clients will speak uTP," he notes. "I suspect this is what is behind the variable effects reported by your users."

Meanwhile, Topolski says ISPs have no real technical excuse to take specific aim at UDP P2P, not least of which being that protocol agnostic throttling systems really don't care. "Being more congestion-responsive than any other Internet use, it cannot congest the last mile," he says. "If some ISP decides they want to buy new updated DPI products to detect and throttle this too, they will be unable to use congestion as the excuse for doing so."

view: topics flat text 
Post a:

Anonymous_
Anonymous
Premium
join:2004-06-21
127.0.0.1
kudos:2
Reviews:
·Comcast
·Time Warner VOIP
·RoadRunner Cable

4 edits

Works Great

i tryed it and it works great i get like 500KB/s to 700KB/s on each torrent download

before i would only get around 150KB/s

But after the torrent Ends the UPLOADING is VERY slow only like 15KB/s

when downloading file the upload is 100KB/s

Matt
All noise, no signal.
Premium
join:2003-07-20
Jamestown, NC
kudos:12

Unable to use Congestion as the Excuse?

said by funchords :
"If some ISP decides they want to buy new updated DPI products to detect and throttle this too, they will be unable to use congestion as the excuse for doing so."
Awe come on Robb, you know some ISPs aren't based in the same reality we are. They can dress any decision to make it fit whatever they want. While they may be unable to use congestion as the excuse when confronted by knowledgeable folks, they'll trumpet it from on-high to the FCC and any other grass-roots AstroTurf agency that will champion their cause to earn a yearly contribution.

Ikyuao

join:2007-02-26
Wichita, KS
Reviews:
·Cox HSI

Re: Unable to use Congestion as the Excuse?

uTorrent with UDP enabled then congestion control is not necessary while being in UDP mode. Since UDP doesn't need congestion control, socket window size buffer, bit flags. That shouldn't have any problems while using UDP socket when new products hits on market for ISP of needing new big router device with updated new features, security, etc.
--
64K TCP WIN is officially dead for long high latency fat network connection across internet.

SimonMorris

@charter.com

approval from:
ThrowDemsOut See Profile

Fixed the quote

quote:
"To be clear, bypassing Bell Canada throttling is not anywhere in the alpha's design motivation. It is just a coincidence. "
-BitTorrent's Simon Morris

ThrowDemsOut
If you can't convince 'em, confuse 'em
Premium
join:2002-03-03
Mullica Hill, NJ
kudos:4

Re: Fixed the quote

said by SimonMorris :

quote:
"To be clear, bypassing Bell Canada throttling is not anywhere in the alpha's design motivation. It is just a coincidence. "
-BitTorrent's Simon Morris
Right!!! It never crossed their mind to change something that was blocking their product across Canada. LOL!!
--
My BLOG .. .. Internet News .. .. My Web Page
Ask yourself one question: 'Do I feel lucky?' Well, do ya punk?

33591094

join:2002-11-19
Canada

1 edit

Re: Fixed the quote

Just like it never crossed bells mind to upgrade their infrastructure instead of reducing services.

Always with the astroturf...
--
All trolls are equal, but A troll is more equal than others.

rawgerz
In Debt we trust
Premium
join:2004-10-03
Grove City, PA

VPN

Find a VPN service and pay a few $ a month to bypass it altogether.

53059959
Temp banned from BBR more then anyone

join:2002-10-02
PwnZone

Re: VPN

As i've proposed in the campus broadband forum a few years back (dead forum now) why don't just a bunch of BBR users chip in together and get an unmetered dedicated server somewhere, use it for VPN.

20mbps unmetered for $140/mo split among 10 people

rawgerz
In Debt we trust
Premium
join:2004-10-03
Grove City, PA

Re: VPN

I once was connected to a seeder that used a ded server. Really fast too. They used these guys
»order.1and1.com/xml/order/MsHost···f=Static
But they seem to be routed through Cogent..
--

You can't make all the people happy all of the time. But it should be common sense to shoot for the majority.

Jerm

join:2000-04-10
Richland, WA
kudos:2
That would work - EXCEPT IT DOESN'T BECAUSE BELL THROTTLES VPN traffic too!

Bell Canada has to be the worst ISP ever...
patcat88

join:2002-04-05
Jamaica, NY
kudos:1

Re: VPN

said by Jerm:

That would work - EXCEPT IT DOESN'T BECAUSE BELL THROTTLES VPN traffic too!

Bell Canada has to be the worst ISP ever...
What about HTTPS?

ThrowDemsOut
If you can't convince 'em, confuse 'em
Premium
join:2002-03-03
Mullica Hill, NJ
kudos:4

Let's not leave out the criticisms in Bennet's rewrite

»www.theregister.co.uk/2008/12/05···ent_udp/
One thing that is certain is that uTP will not reduce the volume of traffic that P2P moves across the internet, something that would be commercial suicide for a company that depends heavily on aggressive file sharers, and pirates, for its popularity.

The overall picture suggests that uTP has a serious flaw. If it simply relies on latency measurements to find preferred paths, it’s likely to favor paths where it’s successfully circumventing management. When a path is managed to give UDP priority over TCP (as is apparently the case in the Bell Canada network,) uTP will see that path as uncongested even as it's struggling to deliver TCP. In this case, uTP will in fact impair other applications, as we suggested in our previous piece.


--
My BLOG .. .. Internet News .. .. My Web Page
Ask yourself one question: 'Do I feel lucky?' Well, do ya punk?
iansltx

join:2007-02-19
Golden, CO
kudos:2

Re: Let's not leave out the criticisms in Bennet's rewrite

I'm sure the makers of TechCruch50 would love your description of BitTorrent. They used Vuze to share the HUGE (8+ GB each) video files for each day of the conference. Worked great, too.

fireflier
Coffee. . .Need Coffee
Premium
join:2001-05-25
Limbo

1 edit
I guess this is the "paradigm break in process" you've been repeatedly spewing eh?

Now the big ISPs will have no choice but to throttle the hell out of users because. . .well, because.

We're waiting for the next telcom talking point to justify traffic management. . .

Please enlighten us.

jig

join:2001-01-05
Hacienda Heights, CA
said by ThrowDemsOut:

»www.theregister.co.uk/2008/12/05···ent_udp/
One thing that is certain is that uTP will not reduce the volume of traffic that P2P moves across the internet, something that would be commercial suicide for a company that depends heavily on aggressive file sharers, and pirates, for its popularity.

The overall picture suggests that uTP has a serious flaw. If it simply relies on latency measurements to find preferred paths, it’s likely to favor paths where it’s successfully circumventing management. When a path is managed to give UDP priority over TCP (as is apparently the case in the Bell Canada network,) uTP will see that path as uncongested even as it's struggling to deliver TCP. In this case, uTP will in fact impair other applications, as we suggested in our previous piece.

he's missing the point. if the management is at the edges of the ISP's network (to reduce their interconnect cost, which is one of their huge issues) then IF uTP is granular enough to reduce throughput PER PEER (rather than globally for all current torrents), the client will automatically gravitate towards intra-ISP peers, which will reduce costs and extra-ISP congestion.

that's if he's right about the client being able to search out lower latency "paths" - (UDP isn't a path, but whatever). if he's not (and i don't think he is), then when it senses any congestion (he even points out that TCP and UDP both add to the overall "congestion"), it should automatically reduce traffic to allow TCP through, whether the TCP is locally generated or not. if anything, management that only affects TCP will hurt the system by creating a virtual congestion needlessly.

by itself, UDP should be easier to "switch" than almost any other protocol, so it should cause the least amount of hardware congestion/MB. this should be a win/win, except that "congestion"'s sliding scale is defined by the makers of utorrent.

it would be really interesting if uTP can throttle per peer. if it can, then the ISPs have reason to cheer - it's an automatic way to keep most of the traffic from crossing over to the backbone.

espaeth
Digital Plumber
Premium,MVM
join:2001-04-21
Minneapolis, MN
kudos:2
Reviews:
·Clear Wireless

Re: Let's not leave out the criticisms in Bennet's rewrite

said by jig:

he's missing the point. if the management is at the edges of the ISP's network (to reduce their interconnect cost, which is one of their huge issues)
I believe that to be a misread of the situation -- the carrier interconnect costs are minimal compared to the rest of the infrastructure costs.

said by jig:

then IF uTP is granular enough to reduce throughput PER PEER (rather than globally for all current torrents), the client will automatically gravitate towards intra-ISP peers, which will reduce costs and extra-ISP congestion.
Even if you sourced 100% of the content from the internal ISP network, you'd have the exact same subscriber access network choke points that you do today.

The problem with doing congestion sensing at the edge is that clients can only determine network impact after it is already happening. In the TCP world solutions like Random Early Discard were implemented to get TCP flows to back off before full link saturation occurs to attempt to limit the overall latency impact across all active flows. The problem is that RED will only discard TCP packets to slow down transfers because of the known guaranteed delivery status of the protocol. UDP gets a pass for a few reasons:

1) There is no layer4 recovery mechanism for drops of UDP traffic
2) Recovery at the application layer is too varied to be identified in fixed-hardware rulesets, as is the case with rigidly defined rulesets with TCP
3) There has been a general assumption that nobody would build a file transfer protocol intended for global use on UDP

Overall I expect with this we'll see history repeat itself as they hit all of the problems that TCP did during its development. What will be new this time around will be seeing the impact this will have on devices that depend on TCP flow information for their daily operations, such as routers doing pre-congestion traffic management and intelligent BGP route reflectors that use TCP flow statistics to alter the metrics used by carriers to select different BGP paths. (ie, monitor for TCP retransmissions to be able to redirect traffic around congested links)

One thing is certain - it will be interesting to see what all of the unanticipated impact will be as a large quantity of uTP traffic gets introduced into the Internet ecosystem.

jig

join:2001-01-05
Hacienda Heights, CA

Re: Let's not leave out the criticisms in Bennet's rewrite

said by espaeth:

said by jig:

he's missing the point. if the management is at the edges of the ISP's network (to reduce their interconnect cost, which is one of their huge issues)
I believe that to be a misread of the situation -- the carrier interconnect costs are minimal compared to the rest of the infrastructure costs.
even if you are right, the infrastructure costs are minimized because the p2p uTP traffic automatically meters itself appropriately IF something isn't already unbalanced in how the system deals with different protocols. regardless of whether you are right about the costs differences, the interconnect costs are what they are choosing to complain about most, lately.

said by jig:

then IF uTP is granular enough to reduce throughput PER PEER (rather than globally for all current torrents), the client will automatically gravitate towards intra-ISP peers, which will reduce costs and extra-ISP congestion.
Even if you sourced 100% of the content from the internal ISP network, you'd have the exact same subscriber access network choke points that you do today.
that's not true unless each DSLAM or cable headnode has unlimited upstream bandwidth. cable might be feeling the bite mostly on upload bandwidth allocated for the full node (250 customers/node?), but that's not the whole story.

as far as the UDP specific stuff goes, UDP is a general encapsulation for custom protocols (and was originally developed as such, including as a testbed for an FTP replacement). rather than talk about what hardware mechanisms there are for UDP control, instead think about what the underlying custom protocol does; in this case, uTP. uTP should take care of congestion issues itself (again, recognizing that "congestion" is built into the protocol or the client application, and not necessarily at the control of the ISP, which may be problematic). finally, have just looked at BGP route reflectors, they seem to be dependent on TCP/IP, but i "think" the initial handshake in uTP is still over TCP. i think once the route is set up, then it can still be used for the UDP traffic. other than that, with the increase in VOIP traffic, ISPs and hardware manufacturers will eventually have to figure out how to manage traffic without relying on TCP alone anyway (or ban all non TCP, which would never be done in the US for any length of time).
--
Catapultam habeo. Nisi pecuniam omnem mihi dabis, ad caput tuum saxum immane mittam.
nasadude

join:2001-10-05
Rockville, MD
Reviews:
·Verizon FiOS
"One thing that is certain is that uTP will not reduce the volume of traffic that P2P moves across the internet,.."

so what? that was not the point in developing this anyway. The implicit suggestion in this statement is how P2P is devouring the internet. In fact, P2P traffic makes up less of total internet b/w than previously, with streaming video and other applications closing fast.

this guy is an incumbent shill; he should be ignored.

maartena
Elmo
Premium
join:2002-05-10
Orange, CA
kudos:1
Reviews:
·AT&T U-Verse
·DIRECTV

Piracy will never die.

No matter how hard the authorities and big-media-industry are trying to make it, there will always be some form of piracy.

10 years ago it was Napster, then Kazaa, now it is Torrents, and in 5 years from now we'll use something else.

And if all else fails, I remember the days when we would spread "warez CD's" at work for $10 or so. (A single CD-R cost about that much back in 1996...) - Nowadays, DVD's go for $10 per 25 or so, and unless you are going to remove these types of media from the market, I think that if online piracy comes to a grinding halt, warez, movies, music, porn and other not-so-legal materials will continue to spread that way.

Instead of trying to fight it, try to make purchasing movies and music cheaper then it is right now. And more importantly, If I want to make a copy of said movie for the DVD player in the car to entertain the kids, or if I want those MP3's on both my iPod, A music CD for the car, AND on the computer at work, I should be allowed and not hindered by DRM.
--
"We've been warned against offering the people of this nation false hope. But in the unlikely story that is America, there has never been anything false about hope" - Barack Obama, 2008.

ThrowDemsOut
If you can't convince 'em, confuse 'em
Premium
join:2002-03-03
Mullica Hill, NJ
kudos:4

Re: Piracy will never die.

said by maartena:

Instead of trying to fight it, try to make purchasing movies and music cheaper then it is right now.
Come on, the price would have to go to zero to satisfy the vast majority of movie thieves. They want it, but don't want to pay ANYTHING for it. If the MPAA dropped the price to 1% of what it is now - most would still steal it because free is cheaper than a $1.
--
My BLOG .. .. Internet News .. .. My Web Page
Ask yourself one question: 'Do I feel lucky?' Well, do ya punk?

Guy Incognito

@teksavvy.com

Re: Piracy will never die.

Price isn't the issue, its how Pirates feel treated and the quality of the product.

If the movie industry started their own usenet style service with full DVDRs and Blu-Rays people would pay for that.

33591094

join:2002-11-19
Canada
It must be true - you're so closely allied with the corporate world that YOU must have the facts.

Only a shill would believe the stuff you spew...
--
All trolls are equal, but A troll is more equal than others.
Kearnstd
Elf Wizard
Premium
join:2002-01-22
Mullica Hill, NJ
the problem is that the MPAA and RIAA aka the MAFIAA want full and total control of media, from purchase to your eyeballs. this model is totally Unacceptable in the digital age, once i buy something it is now mine and it should be my right to use it how i please. the MPAA should not be allowed to weasel the DMCA and say i cant make a copy for my laptop and burn a copy for each DVD player in the house. i own the DVD and it and its contents are now mine.
--
[65 Arcanist]Filan(High Elf) Zone: Broadband Reports

imwhite

@comcast.net

utorrent

There will always be actions taken to bypass throttling/piracy/etc. Its not going to go away, its not going to be "fixed". I salute utorrent for the solution for now.
TechnoScott
Premium
join:2003-03-25
00000

Since when?

Since when did FunChords become a network expert? Does a high post count qualify? When I see his Comcast Network Engineering badge then I might agree that he is a network expert.
BubbaDude

join:2008-06-06
Livermore, CA

Re: Since when?

I generally refer to my lovable friend Robb as a "file sharing enthusiast." But that's just to rile him up. He actually had some sort of Microsoft credential at some point in his life.

battleop

join:2005-09-28
00000

More reasons for ISPs to use caps.

Yet more ammo for the bean counters. Keep evading their traffic shaping and they will put something in you can't evade, caps.

MxxCon

join:1999-11-19
Brooklyn, NY

1 edit

Credit where Credit is due

why Robb Topolski always gets the credit for detecting spoofed RST packets on Comcast when i detected it more than 3 years earlier on OOL »Re: Why Is OOL Selectively Blocking Bit Torrent Pa

I even contacted all the major bittorrent developers to work out a solution
»Good news
»more good news for us
»Re: Why Is OOL Selectively Blocking Bit Torrent Pa



Give credit where credit is due
--
Check out my awesome city of MxxTopia »mxxtopia.myminicity.com/ind or »mxxtopia.myminicity.com (the more people visit, the bigger it is)
BubbaDude

join:2008-06-06
Livermore, CA

Re: Credit where Credit is due

I'll gladly give you credit, but I'd have to know your name.

funchords
Hello
Premium,MVM
join:2001-03-11
Yarmouth Port, MA
kudos:5

3 edits
said by MxxCon:

why Robb Topolski always gets the credit for detecting spoofed RST packets on Comcast when i detected it more than 3 years earlier on OOL »Re: Why Is OOL Selectively Blocking Bit Torrent Pa
Great job!

There were some other reports here and there, but yours is great because it connected the RST to a particular sequence.

You deserve credit for that. That Cox thread at the top of the thread is a good one, too. That one is a new one on me -- I'd never seen that one before.

said by MxxCon:

I even contacted all the major bittorrent developers to work out a solution
»Good news
»more good news for us
»Re: Why Is OOL Selectively Blocking Bit Torrent Pa



Give credit where credit is due
And it's all yours. I did see some others but nobody that had ruled out the distant-end ISP or attributed it to a specific tool. I'm not sure I read your original report before, but your report is especially good. I also wasn't the first one to notice some odd behavior on Comcast that now -- in retrospect, we can say was probably Sandvine. If I didn't figure this out, eventually someone else would have. I was just the first (that I know of) that put it all together involving Comcast/Sandvine and (like you did) the protocol specifics.

I also approached the situation in a similar way as you, with just a few differences. Further, the biggest difference -- the explosive blog/press coverage and Comcast's PR and legal hijinx -- I had nothing to do with. That controversy is what gave the story some real legs.

I was the first to attribute RSTs to Comcast using Sandvine with the sequence being the injected packet right after the distant peer requested file data (ED2K/BitTorrent). (Like you, I also saw it after the BitTorrent bitfield.)

I also tested end-to-end capturing both sides, and I also tested on a different ISP under exactly the same conditions and showed it didn't happen. I also used my real name (which I've always done).

But that didn't make much difference at first. Like yours, when I reported mine, it didn't garner a lot of attention.

I'm really not responsible for what happened next. Three months later, the blogosphere suddenly picked it up (#2 story on Digg that week), then the mainstream press. And even with that, the issue probably would have eventually died if Comcast wasn't such a PR idiot about it.

What's important is that it was fully exposed and it's stopping.

Very good job finding it on OOL. You did a good job and it's cool that you worked with Azureus to work around it. Is it still going on when you're not using lazy bitfield or other obfuscation?
--
Robb Topolski -= funchords.com =- Hillsboro, Oregon
More features, more fun, Join BroadbandReports.com, it's free...

MxxCon

join:1999-11-19
Brooklyn, NY

Re: Credit where Credit is due

I have nothing against you Robb. The publicity you brought to this issue is undeniable.
However, whenever throttling/net neutrality topic comes up, Karl always mentions you as a the 1st documented proof of ISPs messing with the network. I pointed it out to Karl a few times that ISPs have been doing it for alot longer and Comcast wasn't the 1st, but it's fallen on deaf ears.

I don't use my real name because I don't want some of the stupid things i say online to affect me in RL(also similar to the issue i brought up in »Privacy in OOL Direct forum?).

funchords
Hello
Premium,MVM
join:2001-03-11
Yarmouth Port, MA
kudos:5

Re: Credit where Credit is due

said by MxxCon:

I don't use my real name because I don't want some of the stupid things i say online to affect me in RL(also similar to the issue i brought up in »Privacy in OOL Direct forum?).
I know what you mean, I stopped using Comcast for this very same retaliation. They didn't like what I was saying, and attacked me personally -- »Complain about Comcast=Have your Reputation Scrutinized -- regardless of the fact that everything I said was repeatedly and independently shown to be true many months earlier.
--
Robb Topolski -= funchords.com =- Hillsboro, Oregon
More features, more fun, Join BroadbandReports.com, it's free...

Ikyuao

join:2007-02-26
Wichita, KS
Reviews:
·Cox HSI

UDP

uTorrent using socket UDP doesn't prevent ISP throttling issue what if ISP decided to put up a filtering on socket UDP? That will affect on gaming online at same time if UDP is filtered completely by ISP decision.
--
64K TCP WIN is officially dead for long high latency fat network connection across internet.

ctceo
Premium
join:2001-04-26
South Bend, IN

Remember this?

»torrentfreak.com/azureus-bittorr···-page-3/

aefstoggaflm
Open Source Fan
Premium
join:2002-03-04
Bethlehem, PA
kudos:1
Reviews:
·Verizon Online DSL

This is the best solution to aviod ISP blocking & throttling

The best solution (for those of us not forced to buy service from a single incumbent of course) is to vote with your feet - if you catch your ISP blocking inbound or throttling your traffic, switch to one that doesn't. I am sure once their revenue dips, they will reconsider that option.
--
Please use the "yellow (IM) envelope" to contact me and please leave the URL intact.

Friday, 10-Feb 10:36:38 Terms of Use & Privacy | feedback | contact | Hosting by nac.net - DSL,Hosting & Co-lo
over 12.5 years online! © 1999-2012 dslreports.com.