August 1998 - NANOG - lists.nanog.org

RE: Q:Why router with ATM interface comes out earlier than pure SONET interface?
by Christian Kuhtz 04 Aug '98

04 Aug '98

> Hmm. How would you assure per-application Quality-of-Service > without letting > the network explicitely know about application behaviour, i.e. signalling > those parameters and establishing some state within network nodes - read: > connection oriented ? How would you ? Even IP's approaches to this problem > are stateful. Clearly define your requirements. Otherwise this discussion is academic. Give me a scenario and I'll engineer you an IP QoS traffic engineering solution. How many "different" application specific behaviors does one need and make up a manageable quantity of CoS across a service provider network? I don't think you'll ever see more than half a dozen or so anytime soon. How exactly do you sell 64 different classes of service to customers? "I would like CoS #57, please." -- "Would you like fries with that, sir?" How is Cos #57 different from CoS #56 or #58? Signaling assumes connection oriented networking. The question is whether you really need to "signal" across the core of a network. Why does the entire core need to know about everything? IP QoS is fundamentally different from a traffic engineer point. You don't define pipes, like in ATM. You don't associate QoS characteristics to a given pipe, because you don't have pipes like in ATM. You define rates, you define drop policies and queuing policies on a per <insert the a defineable *packet* characteristic here>. My point being that all you ATM QoS bigots need to stop thinking like ATM and forget the idea of a connection pipe at the transporrt layer, and listen to us IP QoS bigots :). Perhaps we actually do have a point, and one doesn't actually need ATM. Am I making any sense? I think a Q&A model for this discussion would be best, especially considering the "bandwidth" limitations of all participating parties.. Cheers, Chris -- Christian Kuhtz, BellSouth Corp., Sr. Network Architect <ck(a)bellsouth.net> 1100 Ashwood Parkway, Atlanta, GA 30338 <ck(a)gnu.org> "Legalize free-enterprise murder: why should governments have all the fun?" -- /usr/games/fortune

2 1

Checked out GIP.ORG lately?
by Michael Dillon 04 Aug '98

04 Aug '98

When was the last time you looked at http://www.gip.org ? Here's a quote from "Internet Foundations: Breaking Technology Bottlenecks": It is to achieve these goals that the Global Internet Project (GIP) - a voluntary, international organization of primarily Internet software, hardware and service providers and telecommunications industry CEOs and senior executives - presents this paper. It aims to identify a set of potentially crippling bottlenecks and offer forward-looking recommendations on how to eliminate or minimize such barriers to sustainable Internet growth. ... The Internet's development to date has been as haphazard as it has been delightful, more a function of random entrepreneurial initiative than of mature planning. The time has come to balance entrepreneurial spirit with sophisticated planning, within the largely voluntary framework that always has characterized Internet growth. ... Toward that end, GIP recommends the convening of a summit, ideally late in 1998 or early in 1999, to explore the foundations of future Internet development. The summit would be international in scope, bringing together key Internet players, at the most senior levels, from both the private and nonprofit sectors to consider opportunities and bottlenecks in the four key areas discussed in this paper - end-user software and hardware, high speed access, backbone networks, and server capabilities. ... Backbone Networks Recommendations * Providers of backbone networks should be able to interconnect with other backbone providers under reasonable terms in a competitive market environment. * Industry should evolve to schemes where services at different grades carry their own costs, dictated by the market. To read the whole paper check out http://www.gip.org/gip10.htm but please do explore the rest of the website as well. -- Michael Dillon - Internet & ISP Consulting Memra Communications Inc. - E-mail: michael(a)memra.com Check the website for my Internet World articles - http://www.memra.com

1 0

Re: OUTAGE: Aug 3, 1998
by Sean Donelan 04 Aug '98

04 Aug '98

It appears several Cascade switches in Worldcom's frame-relay network had a problem. Atlanta, Dallas and Washington DC are reportedly the focus of current repair efforts. Although we saw problems in PVCs passing through several other locations. I received a private comment on my previous message. I'm not trying to pick on any one service provider. I know stuff happens to everyone. I too have royally messed up a network, and have my share of dumb programming mistakes. I attempt to rotate the particular concrete examples used. Hypotheticals are nice, but until you name names, no one pays attention. Which is more effective? Using the Hinsdale Central Office fire as a concrete example, or discussing in general the importance of employee training and responding to remote alarms. There are some common themes which run through my messages. It should be interesting to compare how Worldcom and Ascend handle this incident compared to how AT&T and Cisco handled their frame-relay incident. Oh, by the way, those Worldcom sales people who were calling AT&T customers last time. Guess what. What comes around, goes around. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

1 0

RE: Q:Why router with ATM interface comes out earlier than pure SONET interface?
by Christian Kuhtz 04 Aug '98

04 Aug '98

> On Mon, Aug 03, 1998 at 06:32:35PM -0400, Christian Kuhtz wrote: > > Why not just make ATM variable cell size altogether? > > Try designing switching hardware with bound delays and delay jitter and > you will understand. Implement large-scale switching hardware with > Gigabytes of high-speed memory for buffering and keep costs low. Guys, the above was a tongue in cheek remark. :) Cheers, Chris -- Christian Kuhtz, BellSouth Corp., Sr. Network Architect <ck(a)bellsouth.net> 1100 Ashwood Parkway, Atlanta, GA 30338 <ck(a)gnu.org> "Turnaucka's Law: The attention span of a computer is only as long as its electrical cord." -- /usr/games/fortune

1 0

RE: Q:Why router with ATM interface comes out earlier than pure SONET interface?
by Sean M. Doran 04 Aug '98

04 Aug '98

Here is my sole contribution to this conversation. Christian Kuhtz wrote: | That is, if somebody can figure out a way to do private line (circuit | emulation) through an IP cloud. ATM will be an interim step, in my opinion. Done. Essentially what you do is emulate a connection-oriented network over a connectionless one [jnc bait], and prioritize this emulated connection-oriented service. This entails a design constraint: the aggregate input rate for this emulated CO network to any interface must be less than or equal to the output rate, and traffic shaping must be done to avoid the "bunch up" effect. This also requires a resource reservations mechanism of some nature, ranging from manual configuration coordinated by email/phone to fully dynamic "bandwidth brokerage", and a means of identifying the emulated-circuit traffic at each router where it will receive special treatment in terms of queueing, shaping or both. Please see Van Jacobson, Sally Floyd et al's interesting work on this front at http://www-nrg.ee.lbl.gov/nrg-talks.html and in particular Van Jacobson's presentation "Towards Differentiated Services for the Internet" Also very interesting and related to your question is the research and practical work on Class-Based Queueing at http://www-nrg.ee.lbl.gov/floyd/cbq.html and in particular Dr Floyd's "Notes on CBQ and Guaranteed Service" Van Jacobson also did an excellent NANOG presentation, the video clip of which lives at http://www.nanog.org/mtg-9806/agen0698.html (search down to the RealVideo Broadcast beginning at 10:45 am on Monday, June 8) While you are looking at tasty research notes, please also take a moment to consider the wonderful world of best-efforts services (rather than circuit-emulation) and examine the delicious TCP Friendly page at http://www.psc.edu/networking/tcp_friendly.html and study the paper by Jamshid Mahdavi and Sally Floyd intently. These are among the best things you will read for a while. Please send all your network application developer friends -- especially the multimedia ones -- to that document and to the page in general. The followup paper at http://www-nrg.ee.lbl.gov/floyd/end2end-paper.html and the related links at http://www-nrg.ee.lbl.gov/floyd/tcp_unfriendly.html are also wonderful. Heap scorn on the bad guys, or be prepared to police people's congestion-avoidance. For those of you who don't mind doing some digging and who want to see some practical engineering thinking behind alot of this theoretical stuff, please investigate the IETF diff-serv mailing list. Curtis Villamizar <curtis(a)ans.net> has made some excellent contributions there (from an operator's perspective) that are worth some study, although many others have had useful things to say. (Many others have not, however, but any operator, or anyone else, who reads NANOG is probably pretty good at scanning for content). Ten minutes with Noel Chiappa [the jnc being baited] is also very worthwhile. If you can't arrange that, I believe Gordon Cook did an interview with him in 1997ish that I remember was well worth reading. Finally, the ESNET people have recently done some proofs-of-concept that the equivalent of CBR can be done across a small-i internet and in principle the big-I Internet in the presence of a mix of traffic. This should encourage those people who maintain that the good bits of ATM -- and yes, even I admit that there are some -- can be subsumed into the Internet if and as actually needed. Sean.

1 0

OUTAGE: Aug 3, 1998
by Sean Donelan 04 Aug '98

04 Aug '98

Well, there seems to be something going on. Worldcom is experiencing delays of 30+ minutes answering their customer service lines. Where is the fiber cut or a frame-relay problem this time? Yes, I already asked Worldcom customer service. They didn't know what was happening. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

1 0

Re: Q:Why router with ATM interface comes out earlier than pure SONET interface?
by Tony Li 03 Aug '98

03 Aug '98

| >One other fallacy here: ATM and POS do not provide the same effective | >bandwidth when used as an IP transport. Due to the high encapsulation | >overhead and the sizing overhead of fitting packets into cells, ATM has | >proven to be about 20% less efficient at carrying packets than SONET. | >While this is probably not an issue in a campus or within a building, the | >implications for long lines is enormous. | > | | Okay, so given all the great features that ATM is supposed to have | and the only thing that really sucks about it is the overhead due to the 53 | byte cell size, the obvious question is why can't there be an ATM standard | with, say, 197 ( 4 times the current 48 byte payload) or even 389 ( 8 | times 48 ) byte cells? | Is there something magic about 53 or is the IP over ATM application | still so 'obscure' that there is no interest? Increasing the cell size lowers the efficiency further. 53 is an ATM architectural constant. Change it, and it's no longer ATM. Change it, and you're no longer interoperable. Tony

4 4

Q:Why router with ATM interface comes out earlier than pure SONET interface?
by Yu Ning 03 Aug '98

03 Aug '98

Hi friends, I find in an article, and it says: in 1995,1996 most nationalwide ISP face the demand of higher bandwidth than T3, and at that time the only avalible high bandwidth interface in router in ATM port(say 155M), then this makes many of them choose ATM as their backbone tech. Later, Giga router with POS(Packet over SONET/SDH) 155M(or higher) interface comes out, then the situation is changed. Then my question is: why 155M ATM interface is easier to make than 155M POS interface? Although the net capacity of 155M ATM is less than 155m POS, their capacity for IP packet is really close. I think its nearly the same challenge to pump up a 155M ATM pipe as a 155M POS pipe with IP packet for a router. Thanks for any comments. Yu Ning. _____________________________________________* Yu Ning ATM R&D Center of BUPT (Beijing U. of Posts&Telecom) Beijing (ZIP:100876, MBox:147#), P.R.China mailto:yuning@mindless.com or yuning(a)263.net _____________________________________________

4 4

Re: Q:Why router with ATM interface comes out earlier than pure SONET interface?
by Tony Li 03 Aug '98

03 Aug '98

> I find in an article, and it says: in 1995,1996 most nationalwide ISP > face the demand of higher bandwidth than T3, and at that time the only > avalible high bandwidth interface in router in ATM port(say 155M), then > this makes many of them choose ATM as their backbone tech. Later, Giga > router with POS(Packet over SONET/SDH) 155M(or higher) interface comes > out, then the situation is changed. > > Then my question is: why 155M ATM interface is easier to make than 155M > POS interface? Although the net capacity of 155M ATM is less than 155m > POS, their capacity for IP packet is really close. I think its nearly the > same challenge to pump up a 155M ATM pipe as a 155M POS pipe with IP > packet for a router. The ATM interface is actually harder to make than a POS interface because you have to include the segmentation and reassembly (SAR) functionality in the hardware. Also, the software gets a lot more complicated. Why then did ATM appear first? There was a great deal of publicity given to it. The vision was that it was going to be the 'media' of the future and the hype ran all out of proportion with reality. It took quite a while for the hype to give way, and for engineering and economic realities to truly settle in. And as can be seen from recent announcements, that process is not yet complete. One other fallacy here: ATM and POS do not provide the same effective bandwidth when used as an IP transport. Due to the high encapsulation overhead and the sizing overhead of fitting packets into cells, ATM has proven to be about 20% less efficient at carrying packets than SONET. While this is probably not an issue in a campus or within a building, the implications for long lines is enormous. Tony

2 1

Re: New router technology you'll want to look into
by owen＠DeLong.SJ.CA.US 03 Aug '98

03 Aug '98

> > > Here's some new router technology that I think you'll want to look into. > > > http://www.difference.com/tech.html > > > > route-map set-condiments 10 permit > > Don't forget to account for the encapsulation problem. > Some routers use the CATSUP protocol, some KETCHUP. > > There are roumers that Bay is using something completely > different, of course.... > > Curt- > Yes, while the rest of the world has agreed on the MUSTARD protocol for packet delivery, it appears that the cooks at bay have developed the DIJONAISE protocol. While mostly flavorless, it is very high in fat and overhead. Owen

1 0