[69764] in North American Network Operators' Group
Backbone IP network Economics - peering and transit
daemon@ATHENA.MIT.EDU (Gordon Cook)
Mon Apr 19 21:18:33 2004
Date: Mon, 19 Apr 2004 21:17:43 -0400
To: nanog@merit.edu
From: Gordon Cook <cook@cookreport.com>
Errors-To: owner-nanog-outgoing@merit.edu
Peering? Who needs peering if transit can be had=20
for $20 per megabit per second?
I last had a detailed look at peering and transit=20
economics in the summer of 2002. It is pretty=20
amazing to see what has happened to prices since=20
then. I have a private mail list underway on=20
this subject and published the first part of a=20
two part study on Saturday.
Details available at http://cookreport.com/13.03.shtml
Any opinions here on what MCI coming out of=20
bankruptcy will mean? How will once mighty ATT=20
compete? How can it possibly compete if it=20
actually tries to pay the interest on its bonds?
June 2004 COOK Report page one:
Economic Pressure on Long Haul Fiber
=46ive Years After Bubble Burst Prices Have Plunged=20
Yet Nothing Fundamental Has Been Fixed
Examination of Data Network Woes Shows=20
Termite-Riddled Foundation Leading to More=20
Bankruptcies in Absence of Broader Understanding
Tech-Telcom Recovery, or just a pause before the next round of bankruptcies?
In this issue we explain why we believe that=20
competitive fiber backbones have a failing=20
business model that has driven prices below the=20
cost of maintenance and replacement. We point out=20
that the carriers are making up the difference=20
with cost cutting in every way imaginable and by=20
subsidizing the loosing backbones with what=20
profits remain from voice. As the voice profits=20
disappear via government (GIG BE), corporate, and=20
municipal networks that are buying and lighting=20
their own strands of fiber, and hence leaving the=20
PSTN, another round of bankruptcies looks to be=20
inevitable.
As Eli Noam has warned, when the dominoes start=20
to fall again, the US government, rather than let=20
the telecommunications system collapse, will have=20
no choice but to step in and start regulating IP=20
networks. The kind of regulation that is coming=20
will have little to do with encouraging common=20
carriage or wide spread access to broadband.=20
Instead it will have everything to do with=20
consolidating a few remaining survivors and=20
enabling them to pay their bondholders. The scene=20
is not a pretty one and the miserable policy is=20
being made exacerbated by American bankruptcy law=20
that permits bankrupt fiber carrier to reorganize=20
through Chapter 11 rather than insisting on=20
disbandment via Chapter 7.
What is especially pernicious about this=20
situation is that permitting bankrupt carriers=20
such as MCI to reorganize makes a bad situation=20
worse. It does nothing to get rid of the=20
company's massive glut of fiber. Instead, by=20
wiping out its debt, it ensures a cost cutting=20
advantage to the reorganized company. Still,=20
blessed with the same massive amount of fiber it=20
had before, it can gain income by again cutting=20
prices below what its not-yet-bankrupt cousins=20
like ATT and Sprint can afford to sell, if they=20
are to every pay off their bonds. Roxanne Googin,=20
in an interview to be published next month, was=20
scathing in her comments on MCI being allowed by=20
the Bush Justice Department to file Chapter 11=20
rather than 7 - given the company's acknowledged=20
massive fraud. She added that, given the MCI=20
example, in her opinion any carrier that makes a=20
good faith effort to pay back its bond holders is=20
just plain foolish.
With this issue we turn to the badly broken=20
economics of peering and the attendant backbone=20
business models. After three hours conversation=20
with Farooq Hussain and ninety minutes with=20
=46arooq and Roxane Googin we have a clearer=20
picture of where things went wrong beginning with=20
the privatization of the NSF Net backbone in=20
April 1995. The plan when implemented was a=20
reasonable one, but it is fascinating to look=20
back and understand how all the dominoes fell the=20
wrong way. The result made a bad situation=20
steadily worse until the Cable and Wireless=20
bankruptcy of 2002 ripped a major hole in the=20
Tier One hierarchical fa=E7ade and pointed the way=20
to the technology topology business case issues=20
pointed out by Farooq during our conversation of=20
April 4, 2004.
This fairly short issue lays the foundations for=20
the mail list discussion that began on March 18=20
and will be published in the next issue in about=20
mid May. In focusing on the woes of the LECs=20
during the past two years, we have overlooked the=20
fact that the carriers are in worse shape and=20
that the telcos are again speeding toward the=20
precipice. Most everything is broken and, one of=20
the frustrating issues, is that it is difficult=20
to get agreement on just where. The phone=20
companies do need to die and be replaced. But a=20
huge problem that remains is that, unlike Japan,=20
and many other countries, the US is in gridlock.=20
=46or, in the US, the ILECS and IXCs "own" the FCC.=20
There is still enough money left in the industry=20
that government is hobbled by a political=20
unwillingness to let the bond and equity holders=20
take the consequences and get on with life.
Looking for the Big Picture and By Passing the Carrier
Let's look more closely at what is going on. One=20
consequence of the fiber glut has been the=20
emergence of a large number of ways to send one's=20
"bits."
By this we mean that there are:
(a) At least three distinct technologies used in=20
sending bits over glass - each with its own=20
economics. Layer one technology: OBGP and User=20
Controlled Light Paths technologies from Canarie=20
that are just beginning to be used commercially.=20
Layer two technology - namely Gigabit Ethernet=20
and 10 Gig Ethernet over fiber, sometimes using=20
Sonet. But more often not. Layer three: IP often=20
over ATM using MPLS rather than point-to-point=20
links.
(b) At least two different ways of=20
interconnecting to (carrier pops and at IX's) and=20
that these can be used by customers to ratchet=20
prices down further and further
(c) That high speed Ethernet VLAN and routed VPN=20
services can be bought from carriers, they can=20
also be used as ways to bypass carriers. One by=20
pass business model is to set up a high speed=20
VLAN transport service at numerous exchanges and=20
then sell customers routes over that service.
(d) That outright purchase of dark fiber is also=20
among the ways to by pass carriers - ie=20
corporate, research and education and municipal=20
nets are all doing this.
(e) That for municipal nets there is a new=20
wrinkle being used by Terry McGarty of The Merton=20
Group. Here anyone who puts 20% of the cost down=20
and goes through the proper application=20
procedures can borrow 80% from the Rural Utility=20
Service (RUS) of the United States Department of=20
Agriculture. The RUS fund has 1.2 billion dollars=20
still unspoken for. Describing this business=20
model at David Isenburg's WTF meeting on Saturday=20
April 3, Terry explained how he is moving=20
town-by-town through southern New Hampshire. He=20
is applying a new and potentially powerful model=20
of economic-broker-middleman in taking his=20
consultancy's funds (20%) and getting the US=20
government to match the remaining 80 percent. He=20
shows town management (Hannover for example=20
http://www.hanovernh.org/twn_minbos1-19-04.html )=20
what is possible. He then applies for the funds=20
with the cooperation of the town. When the=20
application is approved, both parties then=20
jointly hire the contractor to build the network.=20
I recorded his talk but stating that he is=20
operating in stealth mode, he so far refuses=20
permission to publish it. I assured him that I am=20
a friend of what he is doing. However, it remains=20
to be seen whether he will come to that=20
conclusion. The essence of what he is doing is=20
certainly clearly on the public record.
Meanwhile=20
http://www.townsfoum.com/HollisNHforum/viewtopic.php?p=3D583=20
describes Terry in Merrimac new Hampshire in=20
January 2003 and www.mertongroup.com/=20
Municipal%20Broadband%20Networks.pdf is a 2002=20
paper laying out the Merton Group concept for=20
building municipal networks. Think about this and=20
hear the "great sucking sound." A question=20
occurs. Once you do (d) and (e) have you doomed=20
LECs as well as carriers?
(f) That we can expect to hear an announcement=20
before the end of the summer of a venture that=20
will begin the role out of broadband wireless=20
connectivity for video, audio, and data
(g) That another carrier PSTN by pass is the=20
global GIG BE net being paid for by the US DoD.
(h) That all owners of fiber are so eager to sell=20
access and get some income that they will do darn=20
near anything to enable attachment, and that=20
Level 3 and Global Crossing may find their=20
infrastructure superfluous in parts of Europe and=20
Asia when they run up against the national PTTs.=20
As will Sprint, MCI and ATT
(i) That profits from voice propping up the Swiss=20
cheese edifice of fiber data nets are fast=20
disappearing - that the center of the PSTN is=20
being hollowed out every which way with wireless=20
and cable waiting in the wings. As the wireline=20
PSTN collapses perhaps cellular will be what=20
remains? It would seem that only in this context=20
could Verizon's amazing offer on April 12, to the=20
=46CC of five billion dollars for ten kilohertz of=20
PCS spectrum make any sense. See=20
http://www.broadbandreports.com/shownews/42022
(j) That the USA is in far worse shape than the rest of the world.
~ ~ ~ ~ ~ ~ ~ ~
Contents
Economic Pressure on Long Haul Fiber
=46ive Years After Bubble Burst Prices Have Plunged=20
Yet Nothing Fundamental Has Been Fixed
Examination of Data Network Woes Shows=20
Termite-Riddled Foundation Leading to More=20
Bankruptcies in Absence of Broader Understanding=20
p. 1
Backbone IP Networks: Why the Hierarchical Peering Model Is Broken
Like Electric Grid - No One Wants to Pay for Connecting
as Fiber Glut Overturns Dominant Position of Tier Ones p. 6
Some of the Business Model Issues of Peering and=20
Interconnection Or Why the Government Could=20
Eventually Be Forced to Take Over p. 19
We Are Now a Decade into IP Based Optical telecom=20
without a Viable Business Model -- An Optical=20
Network Designer Looks at the Collective Insanity=20
of the Bubble and Calls for Fresh Assessment p. 23
Dave Hughes Explains Sip, VoIP and FW Dialup in=20
Voice Communication with Nepal over the Internet=20
p. 26
Digital Photography Comes of Age -- Two New=20
O'Reilly Books Reviewed in the Context of New=20
Tools p. 29
--
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
The COOK Report on Internet Protocol, 431 Greenway Ave, Ewing, NJ 08618 USA
609 882-2572 (PSTN) 703 738-6031 (Vonage)=20
Subscription info & prices at=20
http://cookreport.com/subscriptions.shtml Report=20
on forthcoming backbone bank-
ruptcies at: http://cookreport.com/13.03.shtml E-mail cook@cookreport.com
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
