Interesting People mailing list archives
Re: a wise word from a long time network person -- Merccurynews report on Stanford hearing
From: David Farber <dave () farber net>
Date: Mon, 21 Apr 2008 16:42:39 -0700
________________________________________ From: Tony Lauck [tlauck () madriver com] Sent: Saturday, April 19, 2008 1:48 PM To: David Farber Subject: Re: [IP] a wise word from a long time network person -- Merccurynews report on Stanford hearing There will always be the potential for congestion in *any* shared system that is not grossly over configured. This means there will always be the possibility for congestion in any ISP's network if that ISP has the slightest chance of running a viable business. Therefore, and this is the part where I'm sure Brett and I agree, there will *always* be the necessity to manage congestion in an ISP's network. I have no objection to Comcast's managing its network performance. My objection has been to the *form* of Comcast's management, namely the forging of RST packets. I have also objected to Comcast and others demonizing particular applications protocols or network users. I particularly object to those who criticize the Internet Architecture or IETF without a thorough understanding of the technical issues. I first began working in the area of network congestion management in 1977 when I became chief network architect at Digital Equipment Corporation. In the course of my career at DEC I was instrumental in steering a number of researchers into this area, including Raj Jain and K.K. Ramakrishnan, as well as developing several patents of my own. At the time I told these researchers that this could be a career field if they wanted and not just a project. While many aspects of network performance have become engineering issues, there are still others that are more properly research issues. Because of the complexity of this area, in my opinion the FCC would be ill advised to promulgate regulations that affect congestion management. On the other hand, I would have no problem with the FTC enforcing transparent customer agreements. With dedicated links such as DSL, congestion can, should and is managed at the access multiplexer or router. With dedicated links, congestion appears in the form of a queue inside an intelligent device. At this point, IETF congestion management mechanisms come into play, and performance can be managed by queue discipline and discard policy. However the actual policies are not specified by the IETF, because they are what determine "fair" access. The entire concept of "fair" access depends on what constitutes a "user" and what constitutes "fair" service for that user. This is something that is determined jointly by the ISP and the customer when a customer signs up for network service. All I ask is that these policies be something that ordinary customers as well as network experts can understand. This precludes policies that allow only "reasonable" usage or that disconnect customers for "excessive" usage, without defining these terms. In addition, if usage is limited, than I would expect that the ISP provides customers with simple tools to monitor their usage. These can be similar to the control panel usage monitors provided by shared web hosting companies. As Brett correctly points out, there is at least one other potential bottleneck or cost accumulation point, namely the ISP backbone access link(s). (Depending on geographic considerations, the cost of backbone bandwidth may be more or less significant than last mile costs.) Routers attached to backbone access links can use queue management disciplines to enforce per customer fairness or this can be done at the access router or access multiplexer. Alternatively, backbone access can be monitored and users can be discouraged from excessive usage by usage based tariffs. All I ask is that these charges be open and that the users have a simple way to monitor their usage. Brett has raised a third issue, which is that distributed uploading by P2P networks is inefficient and uneconomic compared with more centralized approaches. This may be true in some instances, particularly with rural networks. However, when looking at the relative costs of multiple approaches it is important to consider *all* the costs involved. These include more than the uplink costs associated with P2P networks. They include the costs associated with uploading data to traditional web and ftp servers, the costs of running these servers and the costs of bandwidth these servers use in sending files. In some cases P2P mechanisms will be more efficient than centralized servers. Two examples come immediately to mind: (1) A home user "publishing" a file that is never accessed. If a centralized server is used there will be a totally unnecessary network transfer uploading the file. (2) A home user sharing an extremely popular file with many other ISP customers. Here the P2P network may reduce the number of copies downloaded over the ISP's backbone access links. I am encouraged by Comcast's newly stated intention to cooperate with Bittorrent. There are significant economies to be realized if all the players cooperate. Unfortunately, there are other factors that may come into play, for example Copyright issues that may prevent ISPs from running their own P2P caching clients. Tony Lauck www.aglauck.com David Farber wrote:
________________________________________ From: Brett Glass [brett () lariat net] Sent: Friday, April 18, 2008 10:06 PM To: David Farber; ip Subject: Re: [IP] a wise word from a long time network person -- Merccurynews report on Stanford hearing At 11:24 AM 4/18/2008, Tony Lauck wrote:Comcast's technical problems are not with the Internet, they are with their DOCSIS 2.0 cable modems, which have limited shared upstream bandwidth and an ineffective multiple access protocol. Presumably these problems will go away when Comcast finally upgrades to DOCSIS 3.0. Other last mile network technologies such as DSL and fiber do not have these problems.This is incorrect. The congestion problems that affect DOCSIS over the neighborhood shared cable affect DSL at the DSLAM. And experience in Japan has demonstrated that adding more bandwidth -- at least up to 100 Mbps per user -- does nothing to satisfy P2P's appetite for bandwidth. In fact, if there is a limit to that appetite, no one can say what it is... because it has never been observed. Also, the problem of cost shifting via P2P is not dependent upon the last mile technology at all. It affects all ISPs in proportion to their upstream bandwidth costs, as I demonstrated during the hearing. I wish I'd had more time to speak. Professor Lessig's talk was disappointing in that it was short on facts and very long on rhetoric. Many of the assertions were unsupported, and there were some ad hominem arguments against Internet providers. (At one point, he likened them to bloodthirsty tigers.) His slides had no graphs, charts, figures, or data from credible sources -- just a few quotes (Gerald Faulhaber was quoted) and words from his talk. In short, I would rate it as a good sermon... but a poor argument. You've heard the old lawyers' saying: "When the law is against you, pound on the facts. When the facts are against you, pound on the law. When both are against you, pound on the table." With all due respect to Larry, whom I admire and who is a truly brilliant lawyer, this particular talk pounded just about entirely on the table. I was potentially a viable opponent, and could have provided a counterargument to every one of Dr. Lessig's points. But the structure of the forum prevented this. Larry rambled for 50 minutes, putting the meeting behind schedule. I spoke as fast as I could for eight very rushed ones. Given the cost of flying from Wyoming, I estimate that I paid at least $100 per minute to speak before the Commissioners -- not counting the two and a half days of work I lost by coming to speak. Larry, on the other hand, was being paid. (As a Stanford professor, he makes more than I do as a rural wireless broadband provider.) Was it worth it? I'm not sure, but when I received the last minute call I realized that I had no choice. I drove to the Denver airport that afternoon through a whiteout snowstorm and flew to California to speak. The text of my prepared remarks, which I wrote on the plane and of which I was able to deliver about half, is at http://www.brettglass.com/FCC on my Web site. As Commissioner Robert McDowell pointed out during the hearing, I was the sole representative of my entire industry who came to speak at the hearing. No one would, or could, speak for me. And my livelihood -- and my 15 years' mission to bring competitive broadband where it never would be available otherwise -- was on the line. If Larry had yielded me merely 5 minutes of the time consumed by hhis speech -- which was full of long, dramatic pauses I couldn't afford to make -- it would have been sufficient for me to make more than a dozen additional points that I did not have the chance to address at the hearing. I would welcome the opportunity to engage Larry in a real, substantive, unhurried debate on this issue. --Brett Glass, Founder and Owner, LARIAT ------------------------------------------- Archives: http://www.listbox.com/member/archive/247/=now RSS Feed: http://www.listbox.com/member/archive/rss/247/ Powered by Listbox: http://www.listbox.com
------------------------------------------- Archives: http://www.listbox.com/member/archive/247/=now RSS Feed: http://www.listbox.com/member/archive/rss/247/ Powered by Listbox: http://www.listbox.com
Current thread:
- a wise word from a long time network person -- Merccurynews report on Stanford hearing David Farber (Apr 18)
- <Possible follow-ups>
- a wise word from a long time network person -- Merccurynews report on Stanford hearing David Farber (Apr 19)
- Re: a wise word from a long time network person -- Merccurynews report on Stanford hearing David Farber (Apr 21)
- Re: a wise word from a long time network person -- Merccurynews report on Stanford hearing David Farber (Apr 22)
- Re: a wise word from a long time network person -- Merccurynews report on Stanford hearing David Farber (Apr 23)