Re: Internet access in Japan (was Re: BitTorrent swarms have a deadly bite on broadband nets)
On Monday 22 October 2007 19:20, David Andersen wrote: > Followed by a recent explosion in fiber-to-the-home buildout by NTT. > "About 8.8 million Japanese homes have fiber lines -- roughly nine > times the number in the United States." -- particularly impressive > when you count that in per-capita terms. Recent? NTT started building the FTC buildout in the mid-90s. At least that's when the plans were first discussed. They took a bold leap back when most people were waffling about WANs and Bellcore was saying SMDS was going to be the way of the future. Now they reap the benefits, while some of us are left behind in the bandwidth ghettos of North America. :-( cheers, --dr -- World Security Pros. Cutting Edge Training, Tools, and Techniques Tokyo, Japan November 29/30 - 2007http://pacsec.jp pgpkey http://dragos.com/ kyxpgp
Re: BitTorrent swarms have a deadly bite on broadband nets
On Tue, 23 Oct 2007 00:35:21 EDT, Sean Donelan said: > This doesn't explain why many universities, most with active, symmetric > ethernet switches in residential dorms, have been deploying packet shaping > technology for even longer than the cable companies. If the answer was > as simple as upgrading everyone to 100Mbps symmetric ethernet, or even > 1Gbps symmetric ethernet, then the university resnet's would be in great > shape. If I didn't know better, I'd say Sean was trolling me, but I'll bite anyhow. ;) Actually, upgrading everybody to 100BaseT makes the problem worse, because then if everybody cranks it up at once, the problem moves from "need upstream links that are $PRICY" into the "need upstream links that are $NOEXIST". We have some 9,000+ students resident on campus. Essentially every single one has a 100BaseT jack, and we're working on getting to Gig-E across the board over the next few years. That leaves us two choices on the upstream side - statistical mux effects (and emulating said effects via traffic shaping), or find a way to trunk 225 40GigE links together. And that's just 9,000 customers - if we were a provider the size of most cable companies, we'd *really* be in trouble. Fortunately, statistical mux effects and a little bit of port-agnostic traffic shaping (you go over a well-publicized upload byte limit for a 24 hour span, you get magically turned into a 56k dialup), we fit quite nicely into a single gig-E link and a 622mbit link. Now if any of you guys have a lead on an affordable way to get 225 40GigE's from here to someplace that can *take* 225 40Gig-E's... ;) pgpD1mqKt7eQo.pgp Description: PGP signature
Re: BitTorrent swarms have a deadly bite on broadband nets
On Tue, 23 Oct 2007, Sean Donelan wrote: Ok, maybe the greedy commercial folks screwed up and deserve what they got; but why are the nobel non-profit universities having the same problems? Because if you look at a residential population with ADSL2+ and 10/10 or 100/100 respectively, the upload/download ratios are reversed, from 1:2 with ADSL2+ (double the amount of download to upload), to 2:1 (double the amount of upload to download). In my experience, the amount of download is approximately the same in both cases, which gives that the upload factor changes 1:4 with the access media symmetry. Otoh, long term savings (several years) on operational costs still make residential ethernet a better deal since experience is that "it just works" as opposed to ADSL2+ where you have a very disturbing signal environment where customers are impacting each other which leads to a lot of customer calls regarding poor quality and varying speeds/bit errors over time. -- Mikael Abrahamssonemail: [EMAIL PROTECTED]
Rack space and bandwidth in Joran
Hi All, I am looking for hosting facilities for about 10-20 racks and Internet transit with good local connectivity in Jordan, can anybody help? Thanks, Leigh Porter UK Broaband/PCCW
Re: The next broadband killer: advanced operating systems?
On Mon, 22 Oct 2007 19:39:48 PDT, Hex Star said: > I can see "advanced operating systems" consuming much more bandwidth > in the near future then is currently the case, especially with the web > 2.0 hype. You obviously have a different concept of "near future" than the rest of us, and you've apparently never been on the pushing end of a software deployment where the pulling end doesn't feel like pulling. I suggest you look at the uptake rate on Vista and various Linux distros and think about how hard it will be to get people to run something *really* different. > the operating system interface will allow it to potentially be > offloaded onto a central server allowing for really quick seamless > deployment of updates and security policies as well as reducing the > necessary size of client machine hard drives. Not only this but it'd I hate to say it, but Microsoft's Patch Tuesday probably *is* already pretty close to "as good as we can make it for real systems". Trying to do *really* seamless updates is a horrorshow, as any refugee from software development for telco switches will testify. (And yes, I spent enough time as a mainframe sysadmin to wish for the days where you'd update once, and all 1,297 online users got the updates at the same time...) Also, the last time I checked, operating systems were growing more slowly than hard drive capacities. So trying to reduce the size is really a fool's errand, unless you're trying to hit a specific size point (for example, once it gets too big to fit on a 700M CD, and you decide to go to DVD, there really is *no* reason to scrimp until you're trying to get it in under 4.7G). You want to make my day? Come up with a way that Joe Sixpack can *back up* that 500 gigabyte hard drive that's in a $600 computer (in other words, if that backup scheme costs Joe much more than $50, *it wont happen*). pgpyhsYt4xMlm.pgp Description: PGP signature
Re: BitTorrent swarms have a deadly bite on broadband nets
On Tue, Oct 23, 2007, Sean Donelan wrote: > > On Mon, 22 Oct 2007, Majdi S. Abbas wrote: > > What hurt these access providers, particularly those in the > >cable market, was a set of failed assumptions. The Internet became a > >commodity, driven by this web thing. As a result, standards like DOCSIS > >developed, and bandwidth was allocated, frequently in an asymmetric > >fashion, to access customers. We have lots of asymmetric access > >technologies, that are not well suited to some new applications. > > This doesn't explain why many universities, most with active, symmetric > ethernet switches in residential dorms, have been deploying packet shaping > technology for even longer than the cable companies. If the answer was > as simple as upgrading everyone to 100Mbps symmetric ethernet, or even > 1Gbps symmetric ethernet, then the university resnet's would be in great > shape. > > Ok, maybe the greedy commercial folks screwed up and deserve what they > got; but why are the nobel non-profit universities having the same > problems? because off the shell p2p stuff doesn't seem to pick up on internal peers behind the great NAT that I've seen dorms behind? :P Adrian
Re: Can P2P applications learn to play fair on networks?
Frank, The problem caching solves in this situation is much less complex than what you are speaking of. Caching toward your client base brings down your transit costs (if you have any)or lowers congestion in congested areas if the solution is installed in the proper place. Caching toward the rest of the world gives you a way to relieve stress on the upstream for sure. Now of course it is a bit outside of the box to think that providers would want to cache not only for their internal customers but also users of the open internet. But realistically that is what they are doing now with any of these peer to peer overlay networks, they just aren't managing the boxes that house the data. Getting it under control and off of problem areas of the network should be the first (and not just future) solution. There are both negative and positive methods of controlling this traffic. We've seen the negative of course, perhaps the positive is to give the user what they want ..just on the providers terms. my 2 cents Rich -- From: "Frank Bulk" <[EMAIL PROTECTED]> Sent: Monday, October 22, 2007 7:42 PM To: "'Rich Groves'" <[EMAIL PROTECTED]>; Subject: RE: Can P2P applications learn to play fair on networks? I don't see how this Oversi caching solution will work with today's HFC deployments -- the demodulation happens in the CMTS, not in the field. And if we're talking about de-coupling the RF from the CMTS, which is what is happening with M-CMTSes (http://broadband.motorola.com/ips/modular_CMTS.html), you're really changing an MSO's architecture. Not that I'm dissing it, as that may be what's necessary to deal with the upstream bandwidth constraint, but that's a future vision, not a current reality. Frank -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Rich Groves Sent: Monday, October 22, 2007 3:06 PM To: nanog@merit.edu Subject: Re: Can P2P applications learn to play fair on networks? I'm a bit late to this conversation but I wanted to throw out a few bits of info not covered. A company called Oversi makes a very interesting solution for caching Torrent and some Kad based overlay networks as well all done through some cool strategically placed taps and prefetching. This way you could "cache out" at whatever rates you want and mark traffic how you wish as well. This does move a statistically significant amount of traffic off of the upstream and on a gigabit ethernet (or something) attached cache server solving large bits of the HFC problem. I am a fan of this method as it does not require a large foot print of inline devices rather a smaller footprint of statics gathering sniffers and caches distributed in places that make sense. Also the people at Bittorrent Inc have a cache discovery protocol so that their clients have the ability to find cache servers with their hashes on them . I am told these methods are in fact covered by the DMCA but remember I am no lawyer. Feel free to reply direct if you want contacts Rich -- From: "Sean Donelan" <[EMAIL PROTECTED]> Sent: Sunday, October 21, 2007 12:24 AM To: Subject: Can P2P applications learn to play fair on networks? Much of the same content is available through NNTP, HTTP and P2P. The content part gets a lot of attention and outrage, but network engineers seem to be responding to something else. If its not the content, why are network engineers at many university networks, enterprise networks, public networks concerned about the impact particular P2P protocols have on network operations? If it was just a single network, maybe they are evil. But when many different networks all start responding, then maybe something else is the problem. The traditional assumption is that all end hosts and applications cooperate and fairly share network resources. NNTP is usually considered a very well-behaved network protocol. Big bandwidth, but sharing network resources. HTTP is a little less behaved, but still roughly seems to share network resources equally with other users. P2P applications seem to be extremely disruptive to other users of shared networks, and causes problems for other "polite" network applications. While it may seem trivial from an academic perspective to do some things, for network engineers the tools are much more limited. User/programmer/etc education doesn't seem to work well. Unless the network enforces a behavor, the rules are often ignored. End users generally can't change how their applications work today even if they wanted too. Putting something in-line across a national/international backbone is extremely difficult. Besides network engineers don't like additional in-line devices, no matter how much the sales people claim its fail-safe. Sampling is easier than monitoring a full network feed. Using netflow sampling or even a SPAN port sampling is
Re: BitTorrent swarms have a deadly bite on broadband nets
On Mon, 22 Oct 2007, Majdi S. Abbas wrote: What hurt these access providers, particularly those in the cable market, was a set of failed assumptions. The Internet became a commodity, driven by this web thing. As a result, standards like DOCSIS developed, and bandwidth was allocated, frequently in an asymmetric fashion, to access customers. We have lots of asymmetric access technologies, that are not well suited to some new applications. This doesn't explain why many universities, most with active, symmetric ethernet switches in residential dorms, have been deploying packet shaping technology for even longer than the cable companies. If the answer was as simple as upgrading everyone to 100Mbps symmetric ethernet, or even 1Gbps symmetric ethernet, then the university resnet's would be in great shape. Ok, maybe the greedy commercial folks screwed up and deserve what they got; but why are the nobel non-profit universities having the same problems?
Re: Internet access in Japan (was Re: BitTorrent swarms have a deadly bite on broadband nets)
Once upon a time, David Andersen <[EMAIL PROTECTED]> said: > But no - I was as happy as everyone else when the CLECs emerged and > provided PRI service at 1/3rd the rate of the ILECs Not only was that CLEC service concetrated in higher-density areas, the PRI prices were often not based in reality. There were a bunch of CLECs with dot.com-style business plans (and they're no longer around). Lucent was practically giving away switches and switch management (and lost big $$$ because of it). CLECs also sold PRIs to ISPs based on reciprocal compensation contracts with the ILECs that were based on incorrect assumptions (that most calls would be from the CLEC to the ILEC); rates based on that were bound to increase as those contracts expired. Back when dialup was king, CLECs selling cheap PRIs to ISPs seemed like a sure-fire way to print money. -- Chris Adams <[EMAIL PROTECTED]> Systems and Network Administrator - HiWAAY Internet Services I don't speak for anybody but myself - that's enough trouble.
Re: Can P2P applications learn to play fair on networks?
Hey Rich. We discussed the technology before but the actual mental click here is important -- thank you. BTW, I *think* it was Randy Bush who said "today's leechers are tomorrow's cachers". His quote was longer but I can't remember it. Gadi. On Mon, 22 Oct 2007, Rich Groves wrote: I'm a bit late to this conversation but I wanted to throw out a few bits of info not covered. A company called Oversi makes a very interesting solution for caching Torrent and some Kad based overlay networks as well all done through some cool strategically placed taps and prefetching. This way you could "cache out" at whatever rates you want and mark traffic how you wish as well. This does move a statistically significant amount of traffic off of the upstream and on a gigabit ethernet (or something) attached cache server solving large bits of the HFC problem. I am a fan of this method as it does not require a large foot print of inline devices rather a smaller footprint of statics gathering sniffers and caches distributed in places that make sense. Also the people at Bittorrent Inc have a cache discovery protocol so that their clients have the ability to find cache servers with their hashes on them . I am told these methods are in fact covered by the DMCA but remember I am no lawyer. Feel free to reply direct if you want contacts Rich -- From: "Sean Donelan" <[EMAIL PROTECTED]> Sent: Sunday, October 21, 2007 12:24 AM To: Subject: Can P2P applications learn to play fair on networks? Much of the same content is available through NNTP, HTTP and P2P. The content part gets a lot of attention and outrage, but network engineers seem to be responding to something else. If its not the content, why are network engineers at many university networks, enterprise networks, public networks concerned about the impact particular P2P protocols have on network operations? If it was just a single network, maybe they are evil. But when many different networks all start responding, then maybe something else is the problem. The traditional assumption is that all end hosts and applications cooperate and fairly share network resources. NNTP is usually considered a very well-behaved network protocol. Big bandwidth, but sharing network resources. HTTP is a little less behaved, but still roughly seems to share network resources equally with other users. P2P applications seem to be extremely disruptive to other users of shared networks, and causes problems for other "polite" network applications. While it may seem trivial from an academic perspective to do some things, for network engineers the tools are much more limited. User/programmer/etc education doesn't seem to work well. Unless the network enforces a behavor, the rules are often ignored. End users generally can't change how their applications work today even if they wanted too. Putting something in-line across a national/international backbone is extremely difficult. Besides network engineers don't like additional in-line devices, no matter how much the sales people claim its fail-safe. Sampling is easier than monitoring a full network feed. Using netflow sampling or even a SPAN port sampling is good enough to detect major issues. For the same reason, asymetric sampling is easier than requiring symetric (or synchronized) sampling. But it also means there will be a limit on the information available to make good and bad decisions. Out-of-band detection limits what controls network engineers can implement on the traffic. USENET has a long history of generating third-party cancel messages. IPS systems and even "passive" taps have long used third-party packets to respond to traffic. DNS servers been used to re-direct subscribers to walled gardens. If applications responded to ICMP Source Quench or other administrative network messages that may be better; but they don't.
Re: Internet access in Japan (was Re: BitTorrent swarms have a deadly bite on broadband nets)
On Oct 22, 2007, at 11:02 PM, Jeff Shultz wrote: David Andersen wrote: http://www.washingtonpost.com/wp-dyn/content/article/2007/08/28/ AR2007082801990.html Followed by a recent explosion in fiber-to-the-home buildout by NTT. "About 8.8 million Japanese homes have fiber lines -- roughly nine times the number in the United States." -- particularly impressive when you count that in per-capita terms. Nice article. Makes you wish... For the days when AT&T ran all the phones? I don't think so... For an environment that encouraged long-term investments with high payoff instead of short term profits. For symmetric 100Mbps residential broadband. But no - I was as happy as everyone else when the CLECs emerged and provided PRI service at 1/3rd the rate of the ILECs, and I really don't care to return to the days of having to rent a telephone from Ma Bell. :) But it's not clear that you can't have both, though doing it in the US with our vastly larger land area is obviously much more difficult. The same thing happened with the CLECs, really -- they provided great, advanced service to customers in major metropolitan areas where the profits were sweet, and left the outlying, low-profit areas to the ILECs. Universal access is a tougher nut to crack. -Dave PGP.sig Description: This is a digitally signed message part
Re: BitTorrent swarms have a deadly bite on broadband nets
According to http://torrentfreak.com/comcast-throttles-bittorrent-traffic-seeding-impossible/ Comcast's blocking affects connections to non-Comcast users. This means that they're trying to manage their upstream connections, not the local loop. For Comcast's own position, see http://bits.blogs.nytimes.com/2007/10/22/comcast-were-delaying-not-blocking-bittorrent-traffic/
Re: Internet access in Japan (was Re: BitTorrent swarms have a deadly bite on broadband nets)
David Andersen wrote: http://www.washingtonpost.com/wp-dyn/content/article/2007/08/28/AR2007082801990.html Followed by a recent explosion in fiber-to-the-home buildout by NTT. "About 8.8 million Japanese homes have fiber lines -- roughly nine times the number in the United States." -- particularly impressive when you count that in per-capita terms. Nice article. Makes you wish... For the days when AT&T ran all the phones? I don't think so...
RE: BitTorrent swarms have a deadly bite on broadband nets
A lot of the MDUs and apartment buildings in Japan are doing fiber to the basement and then VDSL or VDSL2 in the building, or even Ethernet. That's how symmetrical bandwidth is possible. Considering that much of the population does not live in high-rises, this doesn't easily apply to the U.S. population. Frank -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Leo Bicknell Sent: Monday, October 22, 2007 8:55 PM To: nanog@merit.edu Subject: Re: BitTorrent swarms have a deadly bite on broadband nets In a message written on Mon, Oct 22, 2007 at 08:24:17PM -0500, Frank Bulk wrote: > The reality is that copper-based internet access technologies: dial-up, DSL, > and cable modems have made the design-based trade off that there is > substantially more downstream than upstream. With North American > DOCSIS-based cable modem deployments there is generally a 6 MHz wide band at > 256 QAM while the upstream is only 3.2 MHz wide at 16 QAM (or even QPSK). > Even BPON and GPON follow that same asymmetrical track. And the reality is > that most residential internet access patterns reflect that (whether it's a > cause or contributor, I'll let others debate that). Having now seen the cable issue described in technical detail over and over, I have a question. At the most recent Nanog several people talked about 100Mbps symmetric access in Japan for $40 US. This leads me to two questions: 1) Is that accurate? 2) What technology to the use to offer the service at that price point? 3) Is there any chance US providers could offer similar technologies at similar prices, or are there significant differences (regulation, distance etc) that prevent it from being viable? -- Leo Bicknell - [EMAIL PROTECTED] - CCIE 3440 PGP keys at http://www.ufp.org/~bicknell/ Read TMBG List - [EMAIL PROTECTED], www.tmbg.org
RE: Can P2P applications learn to play fair on networks?
I don't see how this Oversi caching solution will work with today's HFC deployments -- the demodulation happens in the CMTS, not in the field. And if we're talking about de-coupling the RF from the CMTS, which is what is happening with M-CMTSes (http://broadband.motorola.com/ips/modular_CMTS.html), you're really changing an MSO's architecture. Not that I'm dissing it, as that may be what's necessary to deal with the upstream bandwidth constraint, but that's a future vision, not a current reality. Frank -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Rich Groves Sent: Monday, October 22, 2007 3:06 PM To: nanog@merit.edu Subject: Re: Can P2P applications learn to play fair on networks? I'm a bit late to this conversation but I wanted to throw out a few bits of info not covered. A company called Oversi makes a very interesting solution for caching Torrent and some Kad based overlay networks as well all done through some cool strategically placed taps and prefetching. This way you could "cache out" at whatever rates you want and mark traffic how you wish as well. This does move a statistically significant amount of traffic off of the upstream and on a gigabit ethernet (or something) attached cache server solving large bits of the HFC problem. I am a fan of this method as it does not require a large foot print of inline devices rather a smaller footprint of statics gathering sniffers and caches distributed in places that make sense. Also the people at Bittorrent Inc have a cache discovery protocol so that their clients have the ability to find cache servers with their hashes on them . I am told these methods are in fact covered by the DMCA but remember I am no lawyer. Feel free to reply direct if you want contacts Rich -- From: "Sean Donelan" <[EMAIL PROTECTED]> Sent: Sunday, October 21, 2007 12:24 AM To: Subject: Can P2P applications learn to play fair on networks? > > Much of the same content is available through NNTP, HTTP and P2P. The > content part gets a lot of attention and outrage, but network engineers > seem to be responding to something else. > > If its not the content, why are network engineers at many university > networks, enterprise networks, public networks concerned about the impact > particular P2P protocols have on network operations? If it was just a > single network, maybe they are evil. But when many different networks > all start responding, then maybe something else is the problem. > > The traditional assumption is that all end hosts and applications > cooperate and fairly share network resources. NNTP is usually considered > a very well-behaved network protocol. Big bandwidth, but sharing network > resources. HTTP is a little less behaved, but still roughly seems to > share network resources equally with other users. P2P applications seem > to be extremely disruptive to other users of shared networks, and causes > problems for other "polite" network applications. > > While it may seem trivial from an academic perspective to do some things, > for network engineers the tools are much more limited. > > User/programmer/etc education doesn't seem to work well. Unless the > network enforces a behavor, the rules are often ignored. End users > generally can't change how their applications work today even if they > wanted too. > > Putting something in-line across a national/international backbone is > extremely difficult. Besides network engineers don't like additional > in-line devices, no matter how much the sales people claim its fail-safe. > > Sampling is easier than monitoring a full network feed. Using netflow > sampling or even a SPAN port sampling is good enough to detect major > issues. For the same reason, asymetric sampling is easier than requiring > symetric (or synchronized) sampling. But it also means there will be > a limit on the information available to make good and bad decisions. > > Out-of-band detection limits what controls network engineers can implement > on the traffic. USENET has a long history of generating third-party cancel > messages. IPS systems and even "passive" taps have long used third-party > packets to respond to traffic. DNS servers been used to re-direct > subscribers to walled gardens. If applications responded to ICMP Source > Quench or other administrative network messages that may be better; but > they don't. > >
Re: The next broadband killer: advanced operating systems?
On 10/21/07, Leo Bicknell <[EMAIL PROTECTED]> wrote: > > Windows Vista, and next week Mac OS X Leopard introduced a significant > improvement to the TCP stack, Window Auto-Tuning. FreeBSD is > committing TCP Socket Buffer Auto-Sizing in FreeBSD 7. I've also > been told similar features are in the 2.6 Kernel used by several > popular Linux distributions. > > Today a large number of consumer / web server combinations are limited > to a 32k window size, which on a 60ms link across the country limits > the speed of a single TCP connection to 533kbytes/sec, or 4.2Mbits/sec. > Users with 6 and 8 MBps broadband connections can't even fill their > pipe on a software download. > > With these improvements in both clients and servers soon these > systems may auto-tune to fill 100Mbps (or larger) pipes. Related > to our current discussion of bittorrent clients as much as they are > "unfair" by trying to use the entire pipe, will these auto-tuning > improvements create the same situation? I can see "advanced operating systems" consuming much more bandwidth in the near future then is currently the case, especially with the web 2.0 hype. In the not so distant future I imagine a operating system whose interface is purely powered by ajax, javascript and some flash with the kernel being a mix of a mozilla engine and the necessary core elements to manage the hardware. This "down to earth" construction of the operating system interface will allow it to potentially be offloaded onto a central server allowing for really quick seamless deployment of updates and security policies as well as reducing the necessary size of client machine hard drives. Not only this but it'd allow the said operating system to easily accept elements from web pages as replacements of core features or additions to already existent features (such as replacing the tray clock with a more advanced clock done in javascript that is on a webpage and whose placement could be done by a simple drag and drop of the code sniplet). Such integration would also open the possibility of applications being made purely of a mixture of various web elements from various webpages. Naturally such a operating environment would be much more intense with regards to its bandwidth consumption requirements but at the same time I can see this as reality in the near future
RE: Can P2P applications learn to play fair on networks?
With PCMM (PacketCable Multimedia, http://www.cedmagazine.com/out-of-the-lab-into-the-wild.aspx) support it's possible to dynamically adjust service flows, as has been done with Comcast's "Powerboost". There also appears to be support for flow prioritization. Regards, Frank -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mikael Abrahamsson Sent: Monday, October 22, 2007 1:02 AM To: nanog@merit.edu Subject: Re: Can P2P applications learn to play fair on networks? On Sun, 21 Oct 2007, Eric Spaeth wrote: > They have. Enter DOCSIS 3.0. The problem is that the benefits of DOCSIS > 3.0 will only come after they've allocated more frequency space, upgraded > their CMTS hardware, upgraded their HFC node hardware where necessary, and > replaced subscriber modems with DOCSIS 3.0 capable versions. On an > optimistic timeline that's at least 18-24 months before things are going to > be better; the problem is things are broken _today_. Could someone who knows DOCSIS 3.0 (perhaps these are general DOCSIS questions) enlighten me (and others?) by responding to a few things I have been thinking about. Let's say cable provider is worried about aggregate upstream capacity for each HFC node that might have a few hundred users. Do the modems support schemes such as "everybody is guaranteed 128 kilobit/s, if there is anything to spare, people can use it but it's marked differently in IP PRECEDENCE and treated accordingly to the HFC node", and then carry it into the IP aggregation layer, where packets could also be treated differently depending on IP PREC. This is in my mind a much better scheme (guarantee subscribers a certain percentage of their total upstream capacity, mark their packets differently if they burst above this), as this is general and not protocol specific. It could of course also differentiate on packet sizes and a lot of other factors. Bad part is that it gives the user an incentive to "hack" their CPE to allow them to send higher speed with high priority traffic, thus hurting their neighbors. -- Mikael Abrahamssonemail: [EMAIL PROTECTED]
RE: Can P2P applications learn to play fair on networks?
Here's a few downstream/upstream numbers and ratios: ADSL2+: 24/1.5 = 16:1 (sans Annex.M) DOCSIS 1.1: 38/9 = 4.2:1 (best case up and downstream modulations and carrier widths) BPON: 622/155 = 4:1 GPON: 2488/1244 = 2:1 Only the first is non-shared, so that even though the ratio is poor, a person can fill their upstream pipe up without impacting their neighbors. It's an interesting question to ask how much engineering decisions have led to the point where we are today with bandwidth-throttling products, or if that would have happened in an entirely symmetrical environment. DOCSIS 2.0 adds support for higher levels of modulation on the upstream, plus wider bandwidth (http://i.cmpnet.com/commsdesign/csd/2002/jun02/imedia-fig1.gif), but still not enough to compensate for the higher downstreams possible with channel bonding in DOCSIS 3.0. Frank -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jack Bates Sent: Monday, October 22, 2007 12:35 PM To: Bora Akyol Cc: Sean Donelan; nanog@merit.edu Subject: Re: Can P2P applications learn to play fair on networks? Bora Akyol wrote: > 1) Legal Liability due to the content being swapped. This is not a technical > matter IMHO. Instead of sending an icmp host unreachable, they are closing the connection via spoofing. I think it's kinder than just dropping the packets all together. > 2) The breakdown of network engineering assumptions that are made when > network operators are designing networks. > > I think network operators that are using boxes like the Sandvine box are > doing this due to (2). This is because P2P traffic hits them where it hurts, > aka the pocketbook. I am sure there are some altruistic network operators > out there, but I would be sincerely surprised if anyone else was concerned > about "fairness" > As has been pointed out a few times, there are issues with CMTS systems, including maximum upstream bandwidth allotted versus maximum downstream bandwidth. I agree that there is an engineering problem, but it is not on the part of network operators. DSL fits in it's own little world, but until VDSL2 was designed, there were hard caps set to down speed versus up speed. This has been how many last mile systems were designed, even in shared bandwidth mediums. More downstream capacity will be needed than upstream. As traffic patterns have changed, the equipment and the standards it is built upon have become antiquated. As a tactical response, many companies do not support the operation of servers for last mile, which has been defined to include p2p seeding. This is their right, and it allows them to protect the precious upstream bandwidth until technology can adapt to a high capacity upstream as well as downstream for the last mile. Currently I show an average 2.5:1-4:1 ratio at each of my pops. Luckily, I run a DSL network. I waste a lot of upstream bandwidth on my backbone. Most downstream/upstream ratios I see on last mile standards and equipment derived from such standards isn't even close to 4:1. I'd expect such ratio's if I filtered out the p2p traffic on my network. If I ran a shared bandwidth last mile system, I'd definitely be filtering unless my overall customer base was small enough to not care about maximums on the CMTS. Fixed downstream/upstream ratios must die in all standards and implementations. It seems a few newer CMTS are moving that direction (though I note one I quickly found mentions it's flexible ratio as beyond DOCSIS 3.0 features which implies the standard is still fixed ratio), but I suspect it will be years before networks can adapt. Jack Bates
Internet access in Japan (was Re: BitTorrent swarms have a deadly bite on broadband nets)
On Oct 22, 2007, at 9:55 PM, Leo Bicknell wrote: Having now seen the cable issue described in technical detail over and over, I have a question. At the most recent Nanog several people talked about 100Mbps symmetric access in Japan for $40 US. This leads me to two questions: 1) Is that accurate? 2) What technology to the use to offer the service at that price point? 3) Is there any chance US providers could offer similar technologies at similar prices, or are there significant differences (regulation, distance etc) that prevent it from being viable? http://www.washingtonpost.com/wp-dyn/content/article/2007/08/28/ AR2007082801990.html The Washington Post article claims that: "Japan has surged ahead of the United States on the wings of better wire and more aggressive government regulation, industry analysts say. The copper wire used to hook up Japanese homes is newer and runs in shorter loops to telephone exchanges than in the United States. ..." a) Dense, urban area (less distance to cover) b) Fresh new wire installed after WWII c) Regulatory environment that forced telecos to provide capacity to Internet providers Followed by a recent explosion in fiber-to-the-home buildout by NTT. "About 8.8 million Japanese homes have fiber lines -- roughly nine times the number in the United States." -- particularly impressive when you count that in per-capita terms. Nice article. Makes you wish... -Dave PGP.sig Description: This is a digitally signed message part
Re: BitTorrent swarms have a deadly bite on broadband nets
In a message written on Mon, Oct 22, 2007 at 08:24:17PM -0500, Frank Bulk wrote: > The reality is that copper-based internet access technologies: dial-up, DSL, > and cable modems have made the design-based trade off that there is > substantially more downstream than upstream. With North American > DOCSIS-based cable modem deployments there is generally a 6 MHz wide band at > 256 QAM while the upstream is only 3.2 MHz wide at 16 QAM (or even QPSK). > Even BPON and GPON follow that same asymmetrical track. And the reality is > that most residential internet access patterns reflect that (whether it's a > cause or contributor, I'll let others debate that). Having now seen the cable issue described in technical detail over and over, I have a question. At the most recent Nanog several people talked about 100Mbps symmetric access in Japan for $40 US. This leads me to two questions: 1) Is that accurate? 2) What technology to the use to offer the service at that price point? 3) Is there any chance US providers could offer similar technologies at similar prices, or are there significant differences (regulation, distance etc) that prevent it from being viable? -- Leo Bicknell - [EMAIL PROTECTED] - CCIE 3440 PGP keys at http://www.ufp.org/~bicknell/ Read TMBG List - [EMAIL PROTECTED], www.tmbg.org pgp1RqRlIg8BG.pgp Description: PGP signature
RE: BitTorrent swarms have a deadly bite on broadband nets
I'm not claiming that squashing P2P is easy, but apparently Comcast has been successfully enough to generate national attention, and the bandwidth shaping providers are not totally a lost cause. The reality is that copper-based internet access technologies: dial-up, DSL, and cable modems have made the design-based trade off that there is substantially more downstream than upstream. With North American DOCSIS-based cable modem deployments there is generally a 6 MHz wide band at 256 QAM while the upstream is only 3.2 MHz wide at 16 QAM (or even QPSK). Even BPON and GPON follow that same asymmetrical track. And the reality is that most residential internet access patterns reflect that (whether it's a cause or contributor, I'll let others debate that). Generally ISPs have been reluctant to pursue usage-based models because it adds an undesirable cost and isn't as attractive a marketing tool to attract customers. Only in business models where bandwidth (local, transport, or otherwise) is expensive has usage-based billing become a reality. Frank -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Crist Clark Sent: Monday, October 22, 2007 7:16 PM To: nanog@merit.edu Subject: RE: BitTorrent swarms have a deadly bite on broadband nets >>> On 10/22/2007 at 3:02 PM, "Frank Bulk" <[EMAIL PROTECTED]> wrote: > I wonder how quickly applications and network gear would implement > QoS support if the major ISPs offered their subscribers two queues: > a default queue, which handled regular internet traffic but > squashed P2P, and then a separate queue that allowed P2P to flow > uninhibited for an extra $5/month, but then ISPs could purchase > cheaper bandwidth for that. > > But perhaps at the end of the day Andrew O. is right and it's best > off to have a single queue and throw more bandwidth at the problem. How does one "squash P2P?" How fast will BitTorrent start hiding it's trivial to spot ".BitTorrent protocol" banner in the handshakes? How many P2P protocols are already blocking/shaping evasive? It seems to me is what hurts the ISPs is the accompanying upload streams, not the download (or at least the ISP feels the same download pain no matter what technology their end user uses to get the data[0]). Throwing more bandwidth does not scale to the number of users we are talking about. Why not suck up and go with the economic solution? Seems like the easy thing is for the ISPs to come clean and admit their "unlimited" service is not and put in upload caps and charge for overages. [0] Or is this maybe P2P's fault only in the sense that it makes so much more content available that there is more for end-users to download now than ever before. B¼information contained in this e-mail message is confidential, intended only for the use of the individual or entity named above. If the reader of this e-mail is not the intended recipient, or the employee or agent responsible to deliver it to the intended recipient, you are hereby notified that any review, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this e-mail in error, please contact [EMAIL PROTECTED]
Re: BitTorrent swarms have a deadly bite on broadband nets
On Mon, Oct 22, 2007 at 05:16:08PM -0700, Crist Clark wrote: > It seems to me is what hurts the ISPs is the accompanying upload > streams, not the download (or at least the ISP feels the same > download pain no matter what technology their end user uses to get > the data[0]). Throwing more bandwidth does not scale to the number > of users we are talking about. Why not suck up and go with the > economic solution? Seems like the easy thing is for the ISPs to come > clean and admit their "unlimited" service is not and put in upload > caps and charge for overages. [I've been trying to stay out of this thread, as I consider it unproductive, but here goes...] What hurts ISPs is not upstream traffic. Most access providers are quite happy with upstream traffic, especially if they manage their upstream caps carefully. Careful management of outbound traffic and an active peer-to-peer customer base, is good for ratios -- something that access providers without large streaming or hosting farms can benefit from. What hurt these access providers, particularly those in the cable market, was a set of failed assumptions. The Internet became a commodity, driven by this web thing. As a result, standards like DOCSIS developed, and bandwidth was allocated, frequently in an asymmetric fashion, to access customers. We have lots of asymmetric access technologies, that are not well suited to some new applications. I cannot honestly say I share Sean's sympathy for Comcast, in this case. I used to work for a fairly notorious provider of co-location services, and I don't recall any great outpouring of sympathy on this list when co-location providers ran out of power and cooling several years ago. I /do/ recall a large number of complaints and the wailing and gnashing of teeth, as well as a lot of discussions at NANOG (both the general session and the hallway track) about the power and cooling situation in general. These have continued through this last year. If the MSOs, their vendors, and our standards bodies in general, have made a failed set of assumptions about traffic ratios and volume in access networks, I don't understand why consumers should be subject to arbitrary changes in policy to cover engineering mistakes. It would be one thing if they simply reduced the upstream caps they offered, it is quite another to actively interfere with some protocols and not others -- if this is truly about upstream capacity, I would expect the former, not the latter. If you read Comcast's services agreement carefully, you'll note that the activity in question isn't mentioned. It only comes up in their Use Policy, something they can and have amended on the fly. It does not appear in the agreement itself. If one were so inclined, one might consider this at least slightly dishonest. Why make a consumer enter into an agreement, which refers to a side agreement, and then update it at will? Can you reasonably expect Joe Sixpack to read and understand what is both a technical and legal document? I would not personally feel comfortable forging RSTs, amending a policy I didn't actually bother to include in my service agreement with my customers, and doing it all to shift the burden for my, or my vendor's engineering assumptions onto my customers -- but perhaps that is why I am an engineer, and not an executive. As an aside, before all these applications become impossible to identify, perhaps it's time for cryptographically authenticated RST cookies? Solving the forging problems might head off everything becoming an encrypted pile of goo on tcp/443. > Information contained in this e-mail message is confidential, intended > only for the use of the individual or entity named above. If the reader > of this e-mail is not the intended recipient, or the employee or agent > responsible to deliver it to the intended recipient, you are hereby > notified that any review, dissemination, distribution or copying of this > communication is strictly prohibited. If you have received this e-mail > in error, please contact [EMAIL PROTECTED] Someone toss this individual a gmail invite...please! --msa
RE: BitTorrent swarms have a deadly bite on broadband nets
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 - -- "Crist Clark" <[EMAIL PROTECTED]> wrote: >[...] How >many P2P protocols are already blocking/shaping evasive? The Storm botnet? :-) - - ferg -BEGIN PGP SIGNATURE- Version: PGP Desktop 9.6.3 (Build 3017) wj8DBQFHHUavq1pz9mNUZTMRAoINAJ4ooF/62eGDSP8ediLys2CifbuUCwCglF/v iPLQgxrMz9iVlHWiUYkfFkQ= =UV+2 -END PGP SIGNATURE- -- "Fergie", a.k.a. Paul Ferguson Engineering Architecture for the Internet fergdawg(at)netzero.net ferg's tech blog: http://fergdawg.blogspot.com/
RE: BitTorrent swarms have a deadly bite on broadband nets
> ... Why not suck up and go with the > economic solution? Seems like the easy thing is for the ISPs to come > clean and admit their "unlimited" service is not and put in upload > caps and charge for overages. Who will be the first? If there *is* competition in the marketplace, the cable company does not want to be the first to say "We limit you" (even if it is true, and has always been true, for some values of truth). This is not a technical problem (telling of the truth), it is a marketing issue. In case it has escaped anyone on this list, I will assert that marketings strengths have never been telling the truth, the whole truth, and nothing but the truth. I read the fine print in my broadband contract. It states that ones mileage (speed) will vary, and the download/upload speeds are maximum only (and lots of other caveats and protections for the provider; none for me, that I recall). But most people do not read the fine contract, but only see the TV advertisements for cable with the turtle, or the flyers in the mail with a cheap price for DSL (so you do not forget, order before midnight tonight!).
RE: BitTorrent swarms have a deadly bite on broadband nets
>>> On 10/22/2007 at 3:02 PM, "Frank Bulk" <[EMAIL PROTECTED]> wrote: > I wonder how quickly applications and network gear would implement QoS > support if the major ISPs offered their subscribers two queues: a default > queue, which handled regular internet traffic but squashed P2P, and then a > separate queue that allowed P2P to flow uninhibited for an extra $5/month, > but then ISPs could purchase cheaper bandwidth for that. > > But perhaps at the end of the day Andrew O. is right and it's best off to > have a single queue and throw more bandwidth at the problem. How does one "squash P2P?" How fast will BitTorrent start hiding it's trivial to spot ".BitTorrent protocol" banner in the handshakes? How many P2P protocols are already blocking/shaping evasive? It seems to me is what hurts the ISPs is the accompanying upload streams, not the download (or at least the ISP feels the same download pain no matter what technology their end user uses to get the data[0]). Throwing more bandwidth does not scale to the number of users we are talking about. Why not suck up and go with the economic solution? Seems like the easy thing is for the ISPs to come clean and admit their "unlimited" service is not and put in upload caps and charge for overages. [0] Or is this maybe P2P's fault only in the sense that it makes so much more content available that there is more for end-users to download now than ever before. B¼information contained in this e-mail message is confidential, intended only for the use of the individual or entity named above. If the reader of this e-mail is not the intended recipient, or the employee or agent responsible to deliver it to the intended recipient, you are hereby notified that any review, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this e-mail in error, please contact [EMAIL PROTECTED]
RE: BitTorrent swarms have a deadly bite on broadband nets
I wonder how quickly applications and network gear would implement QoS support if the major ISPs offered their subscribers two queues: a default queue, which handled regular internet traffic but squashed P2P, and then a separate queue that allowed P2P to flow uninhibited for an extra $5/month, but then ISPs could purchase cheaper bandwidth for that. But perhaps at the end of the day Andrew O. is right and it's best off to have a single queue and throw more bandwidth at the problem. Frank -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Joel Jaeggli Sent: Sunday, October 21, 2007 9:31 PM To: Steven M. Bellovin Cc: Sean Donelan; nanog@merit.edu Subject: Re: BitTorrent swarms have a deadly bite on broadband nets Steven M. Bellovin wrote: > This result is unsurprising and not controversial. TCP achieves > fairness *among flows* because virtually all clients back off in > response to packet drops. BitTorrent, though, uses many flows per > request; furthermore, since its flows are much longer-lived than web or > email, the latter never achieve their full speed even on a per-flow > basis, given TCP's slow-start. The result is fair sharing among > BitTorrent flows, which can only achieve fairness even among BitTorrent > users if they all use the same number of flows per request and have an > even distribution of content that is being uploaded. > > It's always good to measure, but the result here is quite intuitive. > It also supports the notion that some form of traffic engineering is > necessary. The particular point at issue in the current Comcast > situation is not that they do traffic engineering but how they do it. > Dare I say it, it might be somewhat informative to engage in a priority queuing exercise like the Internet-2 scavenger service. In one priority queue goes all the normal traffic and it's allowed to use up to 100% of link capacity, in the other queue goes the traffic you'd like to deliver at lower priority, which given an oversubscribed shared resource on the edge is capped at some percentage of link capacity beyond which performance begins to noticably suffer... when the link is under-utilized low priority traffic can use a significant chunk of it. When high-priority traffic is present it will crowd out the low priority stuff before the link saturates. Now obviously if high priority traffic fills up the link then you have a provisioning issue. I2 characterized this as worst effort service. apps and users could probably be convinced to set dscp bits themselves in exchange for better performance of interactive apps and control traffic vs worst effort services data transfer. Obviously there's room for a discussion of net-neutrality in here someplace. However the closer you do this to the cmts the more likely it is to apply some locally relevant model of fairness. > --Steve Bellovin, http://www.cs.columbia.edu/~smb >
Re: Can P2P applications learn to play fair on networks?
I'm a bit late to this conversation but I wanted to throw out a few bits of info not covered. A company called Oversi makes a very interesting solution for caching Torrent and some Kad based overlay networks as well all done through some cool strategically placed taps and prefetching. This way you could "cache out" at whatever rates you want and mark traffic how you wish as well. This does move a statistically significant amount of traffic off of the upstream and on a gigabit ethernet (or something) attached cache server solving large bits of the HFC problem. I am a fan of this method as it does not require a large foot print of inline devices rather a smaller footprint of statics gathering sniffers and caches distributed in places that make sense. Also the people at Bittorrent Inc have a cache discovery protocol so that their clients have the ability to find cache servers with their hashes on them . I am told these methods are in fact covered by the DMCA but remember I am no lawyer. Feel free to reply direct if you want contacts Rich -- From: "Sean Donelan" <[EMAIL PROTECTED]> Sent: Sunday, October 21, 2007 12:24 AM To: Subject: Can P2P applications learn to play fair on networks? Much of the same content is available through NNTP, HTTP and P2P. The content part gets a lot of attention and outrage, but network engineers seem to be responding to something else. If its not the content, why are network engineers at many university networks, enterprise networks, public networks concerned about the impact particular P2P protocols have on network operations? If it was just a single network, maybe they are evil. But when many different networks all start responding, then maybe something else is the problem. The traditional assumption is that all end hosts and applications cooperate and fairly share network resources. NNTP is usually considered a very well-behaved network protocol. Big bandwidth, but sharing network resources. HTTP is a little less behaved, but still roughly seems to share network resources equally with other users. P2P applications seem to be extremely disruptive to other users of shared networks, and causes problems for other "polite" network applications. While it may seem trivial from an academic perspective to do some things, for network engineers the tools are much more limited. User/programmer/etc education doesn't seem to work well. Unless the network enforces a behavor, the rules are often ignored. End users generally can't change how their applications work today even if they wanted too. Putting something in-line across a national/international backbone is extremely difficult. Besides network engineers don't like additional in-line devices, no matter how much the sales people claim its fail-safe. Sampling is easier than monitoring a full network feed. Using netflow sampling or even a SPAN port sampling is good enough to detect major issues. For the same reason, asymetric sampling is easier than requiring symetric (or synchronized) sampling. But it also means there will be a limit on the information available to make good and bad decisions. Out-of-band detection limits what controls network engineers can implement on the traffic. USENET has a long history of generating third-party cancel messages. IPS systems and even "passive" taps have long used third-party packets to respond to traffic. DNS servers been used to re-direct subscribers to walled gardens. If applications responded to ICMP Source Quench or other administrative network messages that may be better; but they don't.
Re: The next broadband killer: advanced operating systems?
In a message written on Mon, Oct 22, 2007 at 06:42:48PM +0200, Mikael Abrahamsson wrote: > You can achieve the same thing by running a utility such as TCP Optimizer. > > http://www.speedguide.net/downloads.php > > Turn on window scaling and increase the TCP window size to 1 meg or so, > and you should be good to go. A bit of a warning, this is not exactly the same thing. When using the method listed above the system may buffer up to 1 Meg for each active TCP connection. Have 50 people connect to your web server via dialup and the kernel may eat up 50 Meg of memory trying to serve them. That's why the OS defaults have been so low for so long. The auto-tuning method I referenced dynamically changes the size of the window based on the free memory and the speed of the client allowing an individual client to get as big as it needs while insuring fairness. On a single user system with a single TCP connection they both do the same thing. On a very busy web server the first may make it fall over, the second should not. YMMV. -- Leo Bicknell - [EMAIL PROTECTED] - CCIE 3440 PGP keys at http://www.ufp.org/~bicknell/ Read TMBG List - [EMAIL PROTECTED], www.tmbg.org pgpHKVmyrjM6C.pgp Description: PGP signature
Re: Can P2P applications learn to play fair on networks?
Bora Akyol wrote: 1) Legal Liability due to the content being swapped. This is not a technical matter IMHO. Instead of sending an icmp host unreachable, they are closing the connection via spoofing. I think it's kinder than just dropping the packets all together. 2) The breakdown of network engineering assumptions that are made when network operators are designing networks. I think network operators that are using boxes like the Sandvine box are doing this due to (2). This is because P2P traffic hits them where it hurts, aka the pocketbook. I am sure there are some altruistic network operators out there, but I would be sincerely surprised if anyone else was concerned about "fairness" As has been pointed out a few times, there are issues with CMTS systems, including maximum upstream bandwidth allotted versus maximum downstream bandwidth. I agree that there is an engineering problem, but it is not on the part of network operators. DSL fits in it's own little world, but until VDSL2 was designed, there were hard caps set to down speed versus up speed. This has been how many last mile systems were designed, even in shared bandwidth mediums. More downstream capacity will be needed than upstream. As traffic patterns have changed, the equipment and the standards it is built upon have become antiquated. As a tactical response, many companies do not support the operation of servers for last mile, which has been defined to include p2p seeding. This is their right, and it allows them to protect the precious upstream bandwidth until technology can adapt to a high capacity upstream as well as downstream for the last mile. Currently I show an average 2.5:1-4:1 ratio at each of my pops. Luckily, I run a DSL network. I waste a lot of upstream bandwidth on my backbone. Most downstream/upstream ratios I see on last mile standards and equipment derived from such standards isn't even close to 4:1. I'd expect such ratio's if I filtered out the p2p traffic on my network. If I ran a shared bandwidth last mile system, I'd definitely be filtering unless my overall customer base was small enough to not care about maximums on the CMTS. Fixed downstream/upstream ratios must die in all standards and implementations. It seems a few newer CMTS are moving that direction (though I note one I quickly found mentions it's flexible ratio as beyond DOCSIS 3.0 features which implies the standard is still fixed ratio), but I suspect it will be years before networks can adapt. Jack Bates
Re: Can P2P applications learn to play fair on networks?
I see your point. The main problem I see with the traffic shaping or worse boxes is that comcast/ATT/... Sells a particular bandwidth to the customer. Clearly, they don't provision their network as Number_Customers*Data_Rate, they provision it to a data rate capability that is much less than the maximum possible demand. This is where the friction in traffic that you mention below happens. I have to go check on my broadband service contract to see how they word the bandwidth clause. Bora On 10/22/07 9:12 AM, "Sean Donelan" <[EMAIL PROTECTED]> wrote: > On Mon, 22 Oct 2007, Bora Akyol wrote: >> I think network operators that are using boxes like the Sandvine box are >> doing this due to (2). This is because P2P traffic hits them where it hurts, >> aka the pocketbook. I am sure there are some altruistic network operators >> out there, but I would be sincerely surprised if anyone else was concerned >> about "fairness" > > The problem with words is all the good ones are taken. The word > "Fairness" has some excess baggage, nevertheless it is the word used. > > Network operators probably aren't operating from altruistic principles, > but for most network operators when the pain isn't spread equally across > the the customer base it represents a "fairness" issue. If 490 customers > are complaining about bad network performance and the cause is traced to > what 10 customers are doing, the reaction is to hammer the nails sticking > out. > > Whose traffic is more "important?" World of Warcraft lagged or P2P > throttled? The network operator makes P2P a little worse and makes WoW a > little better, and in the end do they end up somewhat "fairly" using the > same network resources. Or do we just put two extremely vocal groups, the > gamers and the p2ps in a locked room and let the death match decide the > winnner?
Re: The next broadband killer: advanced operating systems?
On Mon, 22 Oct 2007, Sam Stickland wrote: Does anyone know if there are any plans by Microsoft to push this out as a Windows XP update as well? You can achieve the same thing by running a utility such as TCP Optimizer. http://www.speedguide.net/downloads.php Turn on window scaling and increase the TCP window size to 1 meg or so, and you should be good to go. The "only" thing this changes for ISPs is that all of a sudden increasing the latency by 30-50ms by buffering in a router that has a link that is full, won't help much, end user machines will be able to cope with that and still use the bw. So if you want to make the gamers happy you might want to look into that WRED drop profile one more time with this in mind if you're in the habit of congesting your core regularily. -- Mikael Abrahamssonemail: [EMAIL PROTECTED]
Re: Can P2P applications learn to play fair on networks?
On Mon, 22 Oct 2007, Bora Akyol wrote: I think network operators that are using boxes like the Sandvine box are doing this due to (2). This is because P2P traffic hits them where it hurts, aka the pocketbook. I am sure there are some altruistic network operators out there, but I would be sincerely surprised if anyone else was concerned about "fairness" The problem with words is all the good ones are taken. The word "Fairness" has some excess baggage, nevertheless it is the word used. Network operators probably aren't operating from altruistic principles, but for most network operators when the pain isn't spread equally across the the customer base it represents a "fairness" issue. If 490 customers are complaining about bad network performance and the cause is traced to what 10 customers are doing, the reaction is to hammer the nails sticking out. Whose traffic is more "important?" World of Warcraft lagged or P2P throttled? The network operator makes P2P a little worse and makes WoW a little better, and in the end do they end up somewhat "fairly" using the same network resources. Or do we just put two extremely vocal groups, the gamers and the p2ps in a locked room and let the death match decide the winnner?
Re: The next broadband killer: advanced operating systems?
Interesting. I imainge this could have a large impact to the typical enterprise, where they might do large scale upgrades in a short period of time. Does anyone know if there are any plans by Microsoft to push this out as a Windows XP update as well? S Leo Bicknell wrote: Windows Vista, and next week Mac OS X Leopard introduced a significant improvement to the TCP stack, Window Auto-Tuning. FreeBSD is committing TCP Socket Buffer Auto-Sizing in FreeBSD 7. I've also been told similar features are in the 2.6 Kernel used by several popular Linux distributions. Today a large number of consumer / web server combinations are limited to a 32k window size, which on a 60ms link across the country limits the speed of a single TCP connection to 533kbytes/sec, or 4.2Mbits/sec. Users with 6 and 8 MBps broadband connections can't even fill their pipe on a software download. With these improvements in both clients and servers soon these systems may auto-tune to fill 100Mbps (or larger) pipes. Related to our current discussion of bittorrent clients as much as they are "unfair" by trying to use the entire pipe, will these auto-tuning improvements create the same situation?
Re: Can P2P applications learn to play fair on networks?
Sean I don't think this is an issue of "fairness." There are two issues at play here: 1) Legal Liability due to the content being swapped. This is not a technical matter IMHO. 2) The breakdown of network engineering assumptions that are made when network operators are designing networks. I think network operators that are using boxes like the Sandvine box are doing this due to (2). This is because P2P traffic hits them where it hurts, aka the pocketbook. I am sure there are some altruistic network operators out there, but I would be sincerely surprised if anyone else was concerned about "fairness" Regards Bora
Re: Can P2P applications learn to play fair on networks?
* Adrian Chadd: > So which ISPs have contributed towards more intelligent p2p content > routing and distribution; stuff which'd play better with their > networks? Perhaps Internet2, with its DC++ hubs? 8-P I think the problem is that better "routing" (Bittorrent content is *not* routed by the protocol AFAIK) inevitably requires software changes. For Bittorrent, you could do something on the tracker side: You serve .torrent files which contain mostly nodes which are topologically close to the requesting IP address. The clients could remain unchanged. (If there's some kind of AS database, you could even mark some nodes as local, so that they only get advertised to nodes within the same AS.) However, there's little incentive for others to use your tracker software. What's worse, it's even less convenient to use because it would need a BGP feed. It's not even obvious if this is going to fix problems. If upload-related congestion on the shared media to the customer is the issue (could be, I don't know), it's unlikely to help to prefer local nodes. It could make things even worse because customers in one area are somewhat likely to be interested in the same data at the same time (for instance, after watching a movie trailer on local TV).
Re: Can P2P applications learn to play fair on networks?
Sean Donelan wrote: Much of the same content is available through NNTP, HTTP and P2P. The content part gets a lot of attention and outrage, but network engineers seem to be responding to something else. If its not the content, why are network engineers at many university networks, enterprise networks, public networks concerned about the impact particular P2P protocols have on network operations? If it was just a single network, maybe they are evil. But when many different networks all start responding, then maybe something else is the problem. The traditional assumption is that all end hosts and applications cooperate and fairly share network resources. NNTP is usually considered a very well-behaved network protocol. Big bandwidth, but sharing network resources. HTTP is a little less behaved, but still roughly seems to share network resources equally with other users. P2P applications seem to be extremely disruptive to other users of shared networks, and causes problems for other "polite" network applications. What exactly is it that P2P applications do that is impolite? AFAIK they are mostly TCP based, so it can't be that they don't have any congestion avoidance, it's just that they utilise multiple TCP flows? Or it is the view that the need for TCP congestion avoidance to kick in is bad in itself (i.e. raw bandwidth consumption)? It seems to me that the problem is more general than just P2P applications, and there are two possible solutions: 1) Some kind of magical quality is given to the network to allow it to do congestion avoidance on an IP basis, rather than on a TCP flow basis. As previously discussed on nanog there are many problems with this approach, not least the fact the core ends up tracking a lot of flow information. 2) A QoS scavenger class is implemented so that users get a guaranteed minimum, with everything above this marked to be dropped first in the event of congestion. Of course, the QoS markings aren't carried inter-provider, but I assume that most of the congestion this thread talks about is occuring the first AS? Sam
NNTP vs P2P (Re: Can P2P applications learn to play fair on networks?)
Adrian Chadd wrote: [..] > Here's the real question. If an open source protocol for p2p content > routing and distribution appeared? It is called NNTP, it exists and is heavily used for doing exactly where most people use P2P for: Warezing around without legal problems. NNTP is of course "nice" to the network as people generally only download, not upload. I don't see the point though, traffic is traffic, somewhere somebody pays for that traffic, from an ISP point of view there is thus no difference between p2p and NNTP. NNTP has quite some overhead (as it is 7bits in general afterall and people need to then encode those 4Gb DVDs ;), but clearly ISPs exist solely for the purpose of providing access to content on NNTP and they are ready to invest lots of money in infrastructure and especially also storage space. I did notice in a recent newsarticle (hardcopy 20min.ch) that the RIAA has finally found NNTP though and are suing Usenet.com though... I wonder what they will do with all those ISPs who are simply selling "NNTP access", who still are claiming that they don't know what they are actually requiring "those big caches" (NNTP servers) for and that they don't know that there is this alt.bin.dvd-r.* stuff on it :) Going to be fun times I guess... Greets, Jeroen signature.asc Description: OpenPGP digital signature
RE: Can P2P applications learn to play fair on networks?
> Would stronger topological sharing be beneficial? If so, how do you > suggest end users software get access to the information required to > make these decisions in an informed manner? I would think simply looking at the TTL of packets from it's peers should be sufficient to decide who is close and who is far away. The problem comes in do you pick someone who is 2 hops away but only has 12K upload or do you pick someone 20 hops away but who has 1M upload? I mean obviously from the point of view of a file sharer, it's speed not location that is important. Geo. George Roettger Netlink Services
Re: Can P2P applications learn to play fair on networks?
On Tue, Oct 23, 2007, Perry Lorier wrote: > Would having a way to proxy p2p downloads via an ISP proxy be used by > ISPs and not abused as an additional way to shutdown and limit p2p > usage? If so how would clients discover these proxies or should they be > manually configured? http://www.azureuswiki.com/index.php/ProxySupport http://www.azureuswiki.com/index.php/JPC Although JPC is now marked "Discontinued due to lack of ISP support." I guess noone wanted to buy their boxes. Would anyone like to see open source JPC-aware P2P caches to build actual meshes inside and between ISPs? Are people even thinking its a good or bad idea? Here's the real question. If an open source protocol for p2p content routing and distribution appeared? The last time I spoke to a few ISPs about it they claimed they didn't want to do it due to possible legal obligations. > Would stronger topological sharing be beneficial? If so, how do you > suggest end users software get access to the information required to > make these decisions in an informed manner? Should p2p clients be > participating in some kind of weird IGP? Should they participate in [snip] As you noted, topological information isn't enough; you need to know about the TE stuff - link capacity, performance, etc. The ISP knows about their network and its current performance much, much more than any edge application would. Unless you're pulling tricks like Cisco OER.. > If p2p clients started using multicast to stream pieces out to peers, > would ISP's make sure that multicast worked (at least within their > AS?). Would this save enough bandwidth for ISP's to care? Can enough > ISP's make use of multicast or would it end up with them hauling the > same data multiple times across their network anyway? Are there any > other obvious ways of getting the bits to the user without them passing > needlessly across the ISP's network several times (often in alternating > directions)? ISPs properly doing multicast pushed from clients? Ahaha. > Should p2p clients set ToS/DSCP/whatever-they're-called-this-week-bits > to state that this is bulk transfers? Would ISP's use these sensibly > or will they just use these hints to add additional barriers into the > network? People who write and the most annoying client users will do whatever they can to maximise their throughput over all others. If this means opening up 50 TCP connections to one host to get the top possible speed and screw the rest of the link, they would. It looks somewhat like GIH's graphs for multi-gige-over-LFN publication.. :) Adrian
Re: Can P2P applications learn to play fair on networks?
Will P2P applications really never learn to play nicely on the network? So from an operations perspective, how should P2P protocols be designed? There appears that the current solution at the moment is for ISP's to put up barriers to P2P usage (like comcasts spoof'd RSTs), and thus P2P clients are trying harder and harder to hide to work around these barriers. Would having a way to proxy p2p downloads via an ISP proxy be used by ISPs and not abused as an additional way to shutdown and limit p2p usage? If so how would clients discover these proxies or should they be manually configured? Would stronger topological sharing be beneficial? If so, how do you suggest end users software get access to the information required to make these decisions in an informed manner? Should p2p clients be participating in some kind of weird IGP? Should they participate in BGP? How can the p2p software understand your TE decisions? At the moment p2p clients upload to a limited number of people, every so often they discard the slowest person and choose someone else. This in theory means that they avoid slow/congested paths for faster ones. Another easy metric they can probably get at is RTT, is RTT a good metric of where operators want traffic to flow? p2p clients can also perhaps do similarity matches based on the remote IP and try and choose people with similar IPs, presumably that isn't going to work well for many people, would it be enough to help significantly? What else should clients be using as metrics for selecting their peers that works in an ISP friendly manner? If p2p clients started using multicast to stream pieces out to peers, would ISP's make sure that multicast worked (at least within their AS?). Would this save enough bandwidth for ISP's to care? Can enough ISP's make use of multicast or would it end up with them hauling the same data multiple times across their network anyway? Are there any other obvious ways of getting the bits to the user without them passing needlessly across the ISP's network several times (often in alternating directions)? Should p2p clients set ToS/DSCP/whatever-they're-called-this-week-bits to state that this is bulk transfers? Would ISP's use these sensibly or will they just use these hints to add additional barriers into the network? Should p2p clients avoid TCP entirely because of it 's "fairness between flows" and try and implement their own congestion control algorithms on top of UDP that attempt to treat all p2p connections as one single "congestion entity"? What happens if this is buggy on the first implementation? Should p2p clients be attempting to mark all their packets as coming from a single application so that ISP's can QoS them as one single entity (eg by setting the IPv6 flowid to the same value for all p2p flows)? What incentive can the ISP provide the end user doing this to keep them from just turning these features off and going back to the current way things are done? Software is easy to fix, and thanks to automatic updates of much p2p network can see a global improvement very quickly. So what other ideas do operations people have for how these things could be fixed from the p2p software point of view?
Re: Can P2P applications learn to play fair on networks?
On Sun, Oct 21, 2007 at 10:45:49PM -0400, Geo. wrote: [snip] > Second, the more people on your network running fileshare network software > and sharing, the less backbone bandwidth your users are going to use when > downloading from a fileshare network because those on your network are > going to supply full bandwidth to them. This means that while your internal > network may see the traffic your expensive backbone connections won't (at > least for the download). Blocking the uploading is a stupid idea because > now all downloading has to come across your backbone connection. As stated in several previous threads on the topic, the clump of p2p protocols in themselves do not provide any topology or locality awareness. At least some of the policing middleboxes have worked with network operators to address the need and bring topology-awareness into varous p2p clouds by eating a BGP feed to redirect traffic on-net (or to non-transit, or same region, or latency class or ...) when possible. Of course the on-net has less long-haul costs, but the last-mile node congestion is killer; at least lower-latency on-net to on-net trafsfers should complete quickly if the network isn't completely hosed. One then can create a token scheme for all the remaining traffic and prioritize, say, the customers actually downloading over those seeding from scratch. -- RSUC / GweepNet / Spunk / FnB / Usenix / SAGE
Re: Can P2P applications learn to play fair on networks?
H... me wonders how you know this for fact? Last time I took the time to snoop a running torrent, I didn't get the the impression it was pulling packets from the same country as I, let alone my network neighbors. That would be totally dependent on what tracker you use. Geo.
Re: BitTorrent swarms have a deadly bite on broadband nets
On Sun, 21 Oct 2007 19:31:09 -0700 Joel Jaeggli <[EMAIL PROTECTED]> wrote: > > Steven M. Bellovin wrote: > > > This result is unsurprising and not controversial. TCP achieves > > fairness *among flows* because virtually all clients back off in > > response to packet drops. BitTorrent, though, uses many flows per > > request; furthermore, since its flows are much longer-lived than web or > > email, the latter never achieve their full speed even on a per-flow > > basis, given TCP's slow-start. The result is fair sharing among > > BitTorrent flows, which can only achieve fairness even among BitTorrent > > users if they all use the same number of flows per request and have an > > even distribution of content that is being uploaded. > > > > It's always good to measure, but the result here is quite intuitive. > > It also supports the notion that some form of traffic engineering is > > necessary. The particular point at issue in the current Comcast > > situation is not that they do traffic engineering but how they do it. > > > > Dare I say it, it might be somewhat informative to engage in a priority > queuing exercise like the Internet-2 scavenger service. > > In one priority queue goes all the normal traffic and it's allowed to > use up to 100% of link capacity, in the other queue goes the traffic > you'd like to deliver at lower priority, which given an oversubscribed > shared resource on the edge is capped at some percentage of link > capacity beyond which performance begins to noticably suffer... when the > link is under-utilized low priority traffic can use a significant chunk > of it. When high-priority traffic is present it will crowd out the low > priority stuff before the link saturates. Now obviously if high priority > traffic fills up the link then you have a provisioning issue. > > I2 characterized this as worst effort service. apps and users could > probably be convinced to set dscp bits themselves in exchange for better > performance of interactive apps and control traffic vs worst effort > services data transfer. > And if you think about these p2p rate limiting devices a bit more broadly, all they really are are traffic classification and QoS policy enforcement devices. If you can set dscp bits with them for certain applications and switch off the policy enforcement feature ... > Obviously there's room for a discussion of net-neutrality in here > someplace. However the closer you do this to the cmts the more likely it > is to apply some locally relevant model of fairness. > > > --Steve Bellovin, http://www.cs.columbia.edu/~smb > > > -- "Sheep are slow and tasty, and therefore must remain constantly alert." - Bruce Schneier, "Beyond Fear"
Re: Can P2P applications learn to play fair on networks?
One of the things to remember is that many customers are simply looking for Internet access, but couldn't tell a megabit from a mackerel. That may have been true 5 years ago, it's not true today. People learn. Here's an interesting issue. I recently learned that the local RR affiliate has changed its service offerings. They now offer 7M/512k resi for $45/mo, or 14M/1M for $50/mo (or thereabouts, prices not exact). Now, does anybody really think that the additional capacity that they're offering for just a few bucks more is real, or are they just playing the numbers for advertising purposes? Windstream offers 6m/384k for $29.95 and 6m/768k for $100, does that answer your question? What is comcast's upspeed, is it this low or is comcast's real problem that they offer 1m or more of upspeed for too cheap a price? Hmmm.. perhaps it's not the customers who don't know a megabit from a mackerel but instead perhaps it's comcast who thinks customers are stupid and as a result they've ended up with the people who want upspeed? Geo. George Roettger Netlink Services
RE: Can P2P applications learn to play fair on networks?
> > > It's a network > > > operations thing... why should Comcast provide a fat pipe for the > > > rest of the world to benefit from? Just my $.02. > > > > Because their customers PAY them to provide that fat pipe? > > You are correct, customers pay Comcast to provide a fat pipe > for THEIR use (MSO's typically understand this as eyeball > heavy content retrieval, not content generation). They do > not provide that pipe for > somebody on another network to use, I mean abuse.Comcast's SLA is > with their user, not the remote user. Comcast is cutting off their user's communication session with a remote user. Since every session on a network involves communications between two customers, only one of whom is usually local, this is the same as randomly killing http sessions or IM sessions or disconnecting voice calls. > Also, its a long standing > policy on most "broadband" type networks that the do not > support user offered services, which this clearly falls into. I agree that there is a bid truth-in-advertising problem here. Cable providers claim to offer Internet access but instead only deliver a Chinese version of the Internet. If you are not sure why I used the term "Chinese", you should do some research on the Great Firewall of China. Ever since the beginning of the commercial Internet, the killer application has been the same. End users want to communicate with other end users. That is what motivates them to pay a monthly fee to an ISP. Any operational measure that interferes with communication is ultimately non-profitable. Currently, it seems that traffic shaping is the least invasive way of limiting the negative impacts. There clearly is demand for P2P file transfer services and there are hundreds of protocols and protocol variations available to do this. We just need to find the right way that meets the needs of both ISPs and end users. To begin with, it helps if ISPs document the technical reasons why P2P protocols impact their networks negatively. Not all networks are built the same. --Michael Dillon
Re: Can P2P applications learn to play fair on networks?
On 10/22/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: > > > It's a network > > operations thing... why should Comcast provide a fat pipe for > > the rest of the world to benefit from? Just my $.02. > > Because their customers PAY them to provide that fat pipe? You are correct, customers pay Comcast to provide a fat pipe for THEIR use (MSO's typically understand this as eyeball heavy content retrieval, not content generation). They do not provide that pipe for somebody on another network to use, I mean abuse.Comcast's SLA is with their user, not the remote user. Also, its a long standing policy on most "broadband" type networks that the do not support user offered services, which this clearly falls into. charles
RE: Can P2P applications learn to play fair on networks?
> So which ISPs have contributed towards more intelligent p2p > content routing and distribution; stuff which'd play better > with their networks? > Or are you all busy being purely reactive? > > Surely one ISP out there has to have investigated ways that > p2p could co-exist with their network.. I can imagine a middlebox that would interrupt multiple flows of the same file, shut off all but one, and then masquerade as the source of the other flows so that everyone still gets their file. If P2P protocols were more transparent, i.e. not port-hopping, this kind of thing would be easier to implement. This would make a good graduate research project, I would imagine. --Michael Dillon
RE: Can P2P applications learn to play fair on networks?
> It's a network > operations thing... why should Comcast provide a fat pipe for > the rest of the world to benefit from? Just my $.02. Because their customers PAY them to provide that fat pipe? --Michael Dillon
[admin] Re: Can P2P applications learn to play fair on networks? and Re: Comcast blocking p2p uploads
On Mon, 22 Oct 2007, Randy Bush wrote: > actually, it would be really helpful to the masses uf us who are being > liberal with our delete keys if someone would summarize the two threads, > comcast p2p management and 204/4. 240/4 has been summarized before: Look for email with "MLC Note" in subject. However, in future, MLC emails will contain "[admin]" in the subject. Interestingly, the content for the p2p threads boils down to: a) Original post by Sean Donelan: Allegation that p2p software "does not play well" with the rest of the network users - unlike TCP-based protocols which results in more or less fair bandwidth allocation, p2p software will monopolize upstream or downstream bandwidth unfairly, resulting in attempts by network operators to control such traffic. Followup by Steve Bellovin noting that if p2p software (like bt) uses tcp-based protocols, due to use of multiple tcp streams, fairness is achieved *between* BT clients, while being unfair to the rest of the network. No relevant discussion of this subject has commenced, which is troubling, as it is, without doubt, very important for network operations. b) Discussion started by Adrian Chadd whether p2p software is aware of network topology or congestion - without apparent answer, which leads me to guess that the answer is "no". c) Offtopic whining about filtering liability, MSO pricing, fairness, equality, end-user complaints about MSOs, filesharing of family photos, disk space provided by MSOs for web hosting. Note: if you find yourself to have posted something that was tossed into the category c) - please reconsider your posting habits. As usual, I apologise if I skipped over your post in this summary. -alex