Below are a variety of publicly available Internet Peering White Papers and Articles.

The documents below tagged with ‘wbn’ are community white papers based on conversations with hundreds of Peering Coordinators, and have gone through a validation process involving walking through the papers with hundreds of Peering Coordinators in public and private fora.

While the price points cited are highly volatile and therefore out-of-date almost immediately, the definitions, the concepts and the logic remain valid.

We also started adding other valuable articles from academic journals, and quality peering resources anywhere we find them.

If you have questions or comments, or would like a walkthrough of any of the papers, please feel free to contact us at ask@drpeering.net.

Other pointers have been added as folks point us to them.

(Editor's Note: Many of these white papers are available across the web at many university sites - this page is storing the authoritative latest versions. Google for "William B. Norton" to find white papers and presentations elsewhere on the net.)

The papers are coded so you know which ones to start with:

Symbol Meaning

These may be of side interest

These are difficult topics, a bit tougher to read through

           

This is by far the most downloaded and discussed white paper on Internet Interconnection, mostly due to the citation by Cringeley.

When faced with the task of distributing video across the Internet, large scale video providers face four choices: a) buy transit, b) buy CDN services, c) buy transit and peer the traffic for free in multiple geographically diverse locations across the U.S., or d) distribute using peer-to-peer methods.  The business model is presented for each, highlighting what ISPs and content providers said were the relevant costs and tradeoffs for equipment, colocation, etc. The implications are significant as the Internet migrates towards being 90% video by the year 2010. Video Internet 1.6 Preso.pdf shared 2/11/09 with Dyyno in Palo Alto. ******

Many large ISPs use Peering Ratios (the ratio of traffic received to traffic sent) as a discriminator for peering candidates. This white paper explores why.  From the title you may infer that most of the arguments for using peering ratios as a discriminator do not seem to hold water. Updated Web version

This is the final in the series of The Business Case for Peering, focusing on the ever decreasing cost of transit and the cost of peering large amounts of traffic using a 10G interface. It takes into account the cost of everything - from transport to equipment to the cost of the interfaces, and yields an interesting result: peering still makes sense, and especially makes sense as traffic volume gets to the point where you need a 10G interface to peer. Updated Web version

This white paper focuses on the definitions and processes of peering. It identifies the 9 selection criteria ISPs told me that they use to determine which IX to build into as well. This one is in use in networking classes at Universities across the US!

I would recommend this one and A Business Case for Peering for folks just getting into peering.  More advanced and more technical folks would be interested in The Art of Peering, as it deals with more strategic "Tricks of the Trade".  Rewritten as Internet Service Providers and Peering (v2.9) web version last updated 2/20/2009 Updated Web version

The Business Case for Peering White Paper is intended for the CFO who just wants the economic argument for peering, specifically, "Why should I spend money on something that won't generate any revenue?" There is after all a cost of peering (transport to an IX, rack and port fees, and equipment costs) and the CFO just wants to be shown the break-even point where peering makes sense. Updated Web version

This white paper demonstrates when *private* peering at an IX makes sense as compared with using point to point circuits. The mathematical models are included, but the price points cited are out of date. *******

This white paper documents the new U.S. Peering Ecosystem that emerged from the ashes of the telecom collapse in 2000-2002. Peering changed dramatically as the Cable companies, large network-savvy content players got into peering in a big way. Updated Web version

This is best suited to broadband and large content providers who want to see if peering fits. They will see the leaders in the industry going down the peering path and saving significant money (>$1 million per annum). The Evolution of the U.S. Peering Ecosystem (v1.2) web version updated 2/23/2008. Updated Web version

This is the latest white paper focusing on the things that Peering Coordinators said they wished they had known before building into Asia. It identifies

5 reasons to build into Asia,

4 International Interconnection Strategies

1 peering dynamic that may seem counter intuitive, along with

9 "lessons learned."

The second half of the paper identifies specific characteristics of the Japan, Hong Kong, Singapore and Australia Peering Ecosystems.

This white paper is most appropriate for those who are exploring expanding into Asia. As the newest, this white paper is still in early draft stage and about to be circulated for broader comments so expect it to be updated several times over the next few months. *****

This is the most controversial peering white paper as it identifies the peering "Tricks of the Trade", those tactics that Peering Coordinators have shared with me that have successfully been used to obtain peering where they otherwise would not have been able to. It is controversial because some of the tactics are ethically questionable - to be listed, the only requirement was that the tactics had to have been successfully used to obtain peering where they otherwise would not have been able to.

This paper is best suited to advanced peering coordinators, those having difficulty obtaining peering (content-heavy ISPs and content players), and technical folks interested in seeing how routing is strategically applied to achieve strategic peering & business goals. Updated Web version

This white paper compares the cost of transit against the cost of peering at ATM-based and Ethernet-Based Internet Exchange points. This was written in 2002 but the results are interesting today; in an environment where customers are comparing transit prices against the cost of transit, rational customers perform the calculations in this paper.

This is a redux of the earlier Business Case for peering, using today’s market prices for peering and transit. Since this is the hottest topic in the Peering Coordinator community, this paper is a must read for anyone that faces the question “Does Peering Sense anymore given that Transit is so Cheap?!”

This virtual simulation game is run in front of a live audience to describe what Internet Peering and Transit are, and how some if the peering dynamics play out.  Four audience members are called to play the role of Peering Coordinators, rolling the dice to acquire network territory, and build into Internet Exchange points where they can negotiate peering. This game takes about 80 minutes and as been run dozens of times at venues around the globe. It has been proven fun and effective at educating the concepts of Internet interconnection. *****

Chang, Jamin, Willinger - this paper explores the peering decision process and the resulting network-specific optimizations of the Internet topology. Changes across the ecosystem are predicted by modeling the decision process and performing a variety of what-if analyses applying a mathematical framework improving upon previous random graph theory analyses.

Dhamdhere and Dovrolis - This paper models the Internet as a set of selfish agents, and explores the profit maximization and utility of the underlying networks taking into account multihoming, geography, and various strategies ISPs emply.

Interconnection and Peering among Internet Service Providers - A Historical perspective, Interisle Consulting Group

This paper provides a thorough and somewhat academic description of Internet interconnections.

When does it make sense to build your own data center and when does it make sense to outsource it?

Consultant referrals are available in a number of geographic locations around the world now - Send e-mail to consultants@DrPeering.net