- Test - lists.nanog.org

More on 'Number of Tiers' Discussion
by Tim Bass 14 Apr '96

14 Apr '96

Food for Thought. Last week or so there was a suggestion here that 'we' adopt a four level tier model for hierarchical routing. I wanted to follow up, just a little, to explain to 'anyone interested' what is the generally accepted 'thinking' on hierarchical routing in large networks today. Kamoun and Kleinrock published an authoritative work on hierarchical routing and came to the interesting discovery: The optimal number of levels (tiers was used recently) is the natural log of the number of routed subnets, or just simply: ln(N) which reduces the routing table to the theoretical optimal number: e*ln(N) If we assume 'the four tier model' as suggested by a poster last week on NANOG, an upper bound is placed on the optimal number of routed subnets in a hierarchical model, basically: 4 = ln(N) or 54 subnets, in the K&K optimal case ;-) Which, BTW would be good for the routing table for an Inter-Hierarchy Routing Protocol (IHRP, for fun :-) .... for 148 entries ! If we, for fun, take the old class of IPv4 address space and simplify the numbers, without worrying about nits; 128 + (191-128)*2^8 + (255-193)*2^16 yields a ballpark number of 123+16128+4063232 ~= 4080 K Hence, ln(4080K) equals 15. So 15 is the optimal number of 'tiers' or levels in the current IPv4 address space (not taking into account distance-path vectors and other fun stuff). Now, of course, we can argue path-length trade offs vs. number-of-levels until our fingers are tired of typing e-mail, but why? The simple idea for illumination in this post is simply; It is not A Good Idea to propose a concrete, structured four-level superstructure for the Internet (and we have not begin to look at IPv6, BTW). This is all just theoretical mumbo-jumbo, BTW. Path length trade-offs, religion, the phase of the moon, greed, genetics, and every other conceivable human condition make ln(N) unobtainable. Our great^N grandchildren will be living in total peace and harmony before humankind will ever see ln(N). But, IMO, 'a four-tier model' is not What We Want, or WWW (note 1). Best Regards, Tim note 1: (a new image for WWW, the old one is getting stale, don't you think?) +------------------------------------------------------------------------+ | Tim Bass | | | Principal Network Systems Engineer | "Never underestimate the bandwidth| | The Silk Road Group, Ltd. | of a station wagon full of tapes | | | hurling down the highway." | | http://www.silkroad.com/ | -Andrew Tanenbaum | | | | +------------------------------------------------------------------------+

2 2

CSPAN at NANOG?
by William B. Norton 12 Apr '96

12 Apr '96

Hi all - Thanks for your feedback regarding CSPAN at the next NANOG. Based on the concerns that some of you raised I've decided *not* to send out the media advisory to CSPAN and the press. Issues leading to that decision included 1) attendees holding their tongues for fear of being misquoted, "Joe X from NetQ said '....' might be stated more succinctly as "NetQ representatives said '.....' or "NetQ said '....' 2) the press misunderstanding and then overemphasizing some issues There was also a fear that marketing folks might be attracted to the bright lights and degrade the technical quality of the meeting. Reasonable concerns. Cheers - Bill ---------------------------------------------------------------------------- ----- William Norton <wbn(a)merit.edu> (313) 936-2656

2 2

Re: CSPAN at NANOG?
by Paul Ferguson 12 Apr '96

12 Apr '96

Thank you. The voice of sanity prevails. :-) - paul At 08:38 AM 4/12/96 -0400, William B. Norton wrote: >Hi all - > >Thanks for your feedback regarding CSPAN at the next NANOG. Based on the >concerns that some of you raised I've decided *not* to send out the media >advisory to CSPAN and the press. Issues leading to that decision included >1) attendees holding their tongues for fear of being misquoted, > "Joe X from NetQ said '....' might be stated more succinctly as > "NetQ representatives said '.....' or "NetQ said '....' >2) the press misunderstanding and then overemphasizing some issues > >There was also a fear that marketing folks might be attracted to the bright >lights and degrade the technical quality of the meeting. Reasonable concerns. > >Cheers - > >Bill >---------------------------------------------------------------------------- >----- >William Norton <wbn(a)merit.edu> (313) 936-2656 > > >

1 0

Re: the Internet Backbone
by Sean Donelan 10 Apr '96

10 Apr '96

> People are out to get smd for the same reason they have been out > to get people throughout history. He's smart, he's right, and he's > moving the world forward. People aren't out to get smd. Most people don't even know or care who smd is. Customers know and care about the results they get. Not who got them, or how many billions was spent getting them. Naturally, its up to each network provider to decide how they want to run their network and their company. If they do an adequate job, their customers will be happy, and their employer and shareholders will be happy. Different providers make different choices running their network. Some providers think they run a better network sending traffic through a few circuitous routes. Other providers think they run a better network using more direct routes. History will tell which providers made the right choice. P.S. I wonder why Mr. Doran thought I was speaking of Sprint when I mentioned complexity for complexities sake. I was actually thinking about the ATM network fiascos. I'm glad to hear talented people at multi-billion dollar companies think only the complexity needed to get the job done was used by talented people at multi-multi-billion dollar companies. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

1 0

Re: the Internet Backbone
by Sean Doran 10 Apr '96

10 Apr '96

| Sean, the problem is that peopel are out to get you. I don't know why this | is, but it is true. It's because I'm an evil asshole determined to protect my employer's interests and make our shareholders rich. This is somewhat incompatible with protecting the interests of our competitors and enriching their shareholders. It is not, however, incompatible with inexpensive access to the Internet through providers who are geared towards offering local support, nor is it incompatible with other clever organizations making money on the Internet too. Personally, I have little patience for the small and not-very-clueful who want to be direct competitors with a multibillion dollar company with lots of talent and who are taken to whining about my policies and those of my colleagues and associates, and even those of our competitors. This uncharitable attitude obviously does not endear me to them. I would hope, though, that the bulk of our customers would be much happier with us driving towards a network reliable enough that they don't have to worry about their customers screaming (not to mention not having to worry about facing some very difficult scaling problems we are already staring at), than with us being the Department of Warm and Fuzzy Feelings. All of them know full well that the drive ain't easy. Sean.

3 3

ISP consolidation
by Srinivasarao Mulugu 10 Apr '96

10 Apr '96

>>>>>>>>>>>>>>>>>>> forwarded message: ... >There are other analysts (myself among them) who anticipate the number of >North Amercan ISP's to grow by a third this year from around 2,000 to >around 3,000. Yes, there is some consolidation, but this is very little >in relation to the entry of new ISP's into the market. ... >>>>>>>>>>>>>>>>>>> forwarded message: ... > I see ISPs going down all the time. Plus, with all the new ones >coming on line, they will not make it like the older times. Unless they >can compete on service. Even though we are $8.50 more in one city, we get >the customers from the other ISP everytime. ... Well, one study I came across talks about a major shake-out over the next few years resulting in about 200 ISPs by 2000. It projects a decline in the number of ISPs beginning early 1997. Of course,some might say, a five-year projection is just about meaningless in a field such as this..... -Mulugu ========================================================= Mulugu Srinivasarao Tel : 703/904-2013 SprintLink Engineering Fax : 703/904-2292 Sprint, GSD Bldg.

4 4

Re: the Internet Backbone
by Sean Doran 10 Apr '96

10 Apr '96

| Some providers "requirements" seem to just add complexity for the | sake of complexity. Oh, please. I have no need to prove the size of my NSPness. The complexity is there because it's needed to make things WORK, so that you can get Internet connectivity. You could help make things easier for everyone to help keep you in business, rather than suggesting that people are out to get you. Sean.

5 4

RE: ISP consolidation
by Jim Fleming 10 Apr '96

10 Apr '96

On Tuesday, April 09, 1996 3:51 PM, Michael Dillon[SMTP:michael@memra.com] wrote: @On Tue, 9 Apr 1996, Srinivasarao Mulugu wrote: @ <snip> @ @> Well, one study I came across talks about a major shake-out over @> the next few years resulting in about 200 ISPs by 2000. It projects @> a decline in the number of ISPs beginning early 1997. @> <snip> @ @That's right. The Internet market and the Internet industry is still @growing by leaps and bounds. In order for a significant shakeout to @occur, that exponential growth curve has to level off. Even then there is @every possibility that there will be plenty of room for the small local ISP @especially when you understand how an ISP operation can be integrated @with a number of other computer/networking oriented businesses. @ @Michael Dillon Voice: +1-604-546-8022 Also, ISPs of the future may not look like ISPs as we know them. For example, I have spoken to churches that want to install a rack of modems and provide dial-up service and e-mail to their congregation. Many of their people are not pleased having their e-mail being spooled at an ISP that browses through their messages. Their church becomes their "ISP". Not only do these churches want to provide traditional services, they also want to use real audio and other voice add-ons to broadcast their services. Special services for seniors are also being considered, such as monitoring for activity in their homes and the dispatching of people to the home if no packets are flowing to check if something is wrong. Another area where ISP growth will occur is in special interest group ISPs. Some groups may choose to provide "free" dial-up service in return for collecting statistics, market research, doing advertising, etc. Also, certain promotions and sweepstakes could be tied to free access via ISPs. For people that only want to use the Internet for a limited set of functions, an ISP could provide free service via a highly subsidized existing business. Stock brokerages and banks are obvious "vendors" that could provide selected customers with free or low-cost access. Some large banks already provide customers with free limos why not free Internet access? No matter what happens, Internet Access will be driven to a commodity pricing level as people and companies discover that they can provide a minimum number of services and still look good because of the synergy of the Internet. This may end if large information providers find that they can market their publications to selected ISPs that are willing to pay a small subscription fee based on the number of users supported by the ISP. This may cause a shake-out if ISPs find that they can not afford the fees and subscribers decide to go where the information flows freely or where truly unique services are offered via ISPs that have an exclusive for the region. -- Jim Fleming UNETY Systems, Inc. Naperville, IL 60563 e-mail: JimFleming(a)unety.net

1 0

Re: the Internet Backbone
by Sean Doran 09 Apr '96

09 Apr '96

Of course, it could also be paranoia, perhaps brought on by the delusion that this was nanog and not com-priv... ;> TFAAML? Sean.

1 0

Re: the Internet Backbone
by Sean Donelan 09 Apr '96

09 Apr '96

Although Sprint requested I not contact its engineers, I'll make and exception in this case. >| Some providers "requirements" seem to just add complexity for the >| sake of complexity. > >Oh, please. I have no need to prove the size of my NSPness. I would appreciate it if you showed me the same courtesy. >The complexity is there because it's needed to make things >WORK, so that you can get Internet connectivity. I'm very aware how well things are working. I have lots and lots of customers that feel things are working so well they've spent their own money to buy separate network lines to St. Louis because of the way things are "working." >You could help make things easier for everyone to help keep >you in business, rather than suggesting that people are out >to get you. The problem is you keep helping stay in this network business, I've been trying to get out of it. We've been trying to make things easier for customers to stay with their current network providers. We installed additional lines on the east coast, west coast, and Canada and offered to peer at additional places so other networks don't need to haul packets more than necessary and hopefully relieve some congestion. Instead some providers choose not to use the alternate paths we provided, but prefer to play hot potato with packets sending them twice around the continent. The result is customers ended up with and average of 31% packet loss between St. Louis and Dallas today. So customers feel the need to buy even more lines to St. Louis. My network business is growing like crazy, but you're killing my real business with your "help." I'm trying to help you keep your customers, why won't you accept the offer? -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

1 0