The Rise of Utility Computing in the Context of the Global Grid

Could it Ever Offer Carriers an Opportunity to Cut the Umbilical to Voice?

We examine utility computing in the context of Google's global grid and the role of the internet as the fundamental infrastructure of the 21st Century. How to purchase this issue. $350 or $1400 group.

March 1, Ewing, NJ -- The April issue looks at utility computing, and at 3Tera with its AppLogic OS wondering if BT in its early committment can build successful 21 CN network business model for enterprise computing as a replacement for reliance on voice.

Executive Summary

Utility Computing, pp: 1-5

This issue examines how increases in processing power and decreases in the cost of storage combined with affordable bandwidth on a global scale are enabling what is variously referred to as grid, utility and cloud computing.

While users still have their edge-based infrastructure and the client server model of the 1990s has, by no means disappeared, the maturation of grid computing into utility computing offers companies of all sizes the opportunity to use the internet as a more powerful and less expensive alternative to client server computing. In many ways with these events, the network is becoming the computer. User’s machines are plugging, like intelligent terminals, into services offered by Amazon and Google and increasingly others that run on top of utility computing infrastructure and offer entrepreneurs ways to build services in software without having to buy and operate complex expensive hardware.


These changes underway at the level of the internet’s “engine rooms” have profound implications. Are we, for example, looking at the emergence of a global computing utility that is becoming as important as the electricity grid? Also does the emergence of utility computing offer the carriers an inducement to transform their networks intro data networks the primary purpose of which is to offer enterprises access to computing services of all types that could be safely and securely performed within the carrier’s cloud rather than in house by IT staff at each separate enterprise. In such a scenario voice is no longer the major carrier product but rather a feature of all the data applications that utility computing enables. We shall see that there is some evidence that this the business strategy behind BT’s 21CN, Global, IP Network’s using 3Tera’s AppLogic Utility Computing OS to run its own applications and then to offer services to BT customers.

(You have just read the introductory paragraphs of page one. The reminder of the introduction. discusses Nicholas Carr’s The Big Switch: Rewiring the World from Edison to Google, Carr shows how Google style computing is becoming an “open source” alternative to costly and proprietary enterprise data centers. Carr explains how all this is made possible by virtualization and then he goes on to discuss 3Tera as the leading example of virtualization enabling utility computing. The introduction then describes how BT has embraced 3Tera for itself and for its enterprise customers. It closes by raising the question of whether carriers can transform themselves by turning themselves from purveyors of voice into distributors of utility computing?

3Tera interview, pp. 6-17

An interview with Barry Lynnn, the CEO and Bert Armijo VP of Sales and Product Management describes the rise of grid cloud and utility computing differentiating between each. We then discuss the origins of 3 tera – a company that resulted from the merger of two independent efforts in the grid computing arena. As they explain what they were doing it was starting out to solve the problem of how to scale online applications.

This led them in a direction where they developed grid tools that enabled applications to be ported to the grid without rewriting code. They called what they came up with AppLogic – an operating system for utility computing.
According to Armijo: AppLogic does not require any unique of specialized infrastructure. It is simply Intel or AMD servers, direct attached storage and gigabit Ethernet. There is no unique hardware whatsoever involved in building an AppLogic grid. We are trying to enable operators – people who already own large numbers of servers to add utility computing services to their product mix.”

AppLogic is available at 17 data centers around the world. It can be used in rebranded form by paying hourly fees to the hosting center. Then they have have small and medium business customers who sign up for the service directly with us. We direct them to hosting partners but we provide first line of support for them.

Most recently, we started offering AppLogic for the enterprise. This is our software with an enterprise-licensing model attached to it. In this case instead of installing it on a metered basis we install it on a server-licensed basis.

Symposium Discussion Mid January to Mid February, p. 26

In EU Regulation as Guarantor of Open Access p. 18

Editor: A society must be able to allocate finite resources for the benefit of its citizens. As I begin to grow more familiar with the long wave thought of Carlota Perez this strikes me as symptomatic of what she calls the finance capital wave. We can’t agree on a course of action because we are still under the grip of what she calls the installation wave. Although we are in the transition between installation and deployment what is possible is still being constricted by the demands of finance capital. To the extent I understand the Perezian point of view we must shift to a production capital mindset and the most likely way of accomplishing that is for the government to step in and take a larger role again. But what she calls intelligent government working in a core shared vision with business and society. I confess fascination as I am listening to her San Francisco talk Feb 21, 2008 and earlier today to her Amsterdam talk from 2007.

EU-commissioner Viviane Reding (speech of January 16) “The regulation of access networks keeps open the potential bottleneck to competition in broadband markets. European rules have provided a real choice for consumers.”
“The European model is empirically proven to promote not just choice, competition and innovation but also investment. Redding goes on to say that European stimulus of investment has been very successful

Jim Kayne: It's not as simple as "build it and they will come". Rather, it's the ability to generate flow. Impeded flow (read scarcity) stifles growth. You can't tell what will develop; whether it will be replicative or novel; primary, secondary, or ternary; interactive locally, globally or in-between.

On the one hand, if the construction is funded by bonds or at least with municipal backing, the risk of default is tiny, so the banks get their money no matter what.

On the other hand, if the project becomes a disaster, then it is the taxpayers who get hosed.

On the other other hand, irrespective of how it's funded - private or muni - the lower you can drive down the cost/home the better off you are. So an offhand assumption of $2,000 will really hurt you if in the real world it's $1,000 or $4,000.

Just some random observations at this point... snip

Vint Cerf: This strikes me as an argument for government-sponsored infrastructure that does NOT seek high rates of return but seeks compensation to recover maintenance costs. The interstate highway system generally falls into a similar category unless I am over- simplifying. snip

COOK Report: Hah!

I agree with you Vint. But has not fealty to the Chicago School been such that this would impossible to do? I mean if the "gummit" meddles in the free market its a fate worse than death!@!! (to the Chicago boys). snip

Waclawsky: Some people I have spoken to in the industry have given me estimates of network cost anywhere from 5% to 30% for bearer infrastructure and of course the rest is for the control/billing/management/etc. infrastructure. It seems that all the effort to monetize all the potential uses of the asset (connectivity), is much more costly (and complex) than the asset itself.
Conal Henry: Richer communications, a wider field of experience, richer access to the rest of the world, improved communication and information management - all things the consumer will enjoy when they have them but not things they tend to miss when they don't.

Wireless p. 29

Very significant wireless developments on the horizon. Harold Feld: Sascha Meinrath (who is also on this list) has put together some "notes from the field" for early deployment in the 3.65 GHz band. Given that this band was only opened 6 months ago, this is very promising for deployment. You can see Sascha's posting:

Peter Ecclesine on 802.11y: "As Wi-Fi radios evolve, they tune more widely into adjacent spectrum, like from 5 GHz to the 4.9 GHz Public Safety spectrum and the 5.9 GHz Intelligent Transportation System spectrum. What has been missing is the software/firmware for licensed operators to control the radios.

These radios running under the FCC lightly licensed scheme will use an over the air control system; listening for a local infrastructure signal that enables them to negotiate link and function responses. They don't operate until they receive an enabling signal to begin operation.

Peter and Sascha will be at the cook-in on march 24 discussing a path involving the 3650 space, 802.11y and the emergence of optical photonics that may cause the difference between wireless and fiber to almost disappear.
Editor: More problem with the finance capital installation era
Tim Cowen (lead counsel BT): The issue with AT+T is that the monopoly area is not well defined, not well regulated and the risk of ability to act as gatekeeper or extract monopoly rents it is increased. But that is what you would expect to get with an ineffectively regulated monopolist.

Anyone disagree?

Goldstein: That's the point I'm making. In a competitive market, their plans would go nowhere fast. As I wrote in mid-2005

(, the "fat wasteband" scenario of DPI'd "broadband" would never fly if there were competition at the IP layer, and therefore the big ILECs have demanded (and gotten) forbearance from common carrier obligations. Thus there is no longer an independent ISP function. The wire owner controls the content. That's an ineffectively regulated monopolist.

An IP Economy Lesson from Kevin Kelly p. 35

Ed Pimentel: I suggest a look at

Editor: Good recommendation from Ed. Here are a few snippets:

Better Than Free

The internet is a copy machine. At its most foundational level, it copies every action, every character, every thought we make while we ride upon it. In order to send a message from one corner of the internet to another, the protocols of communication demand that the whole message be copied along the way several times. IT companies make a lot of money selling equipment that facilitates this ceaseless copying. Every bit of data ever produced on any computer is copied somewhere. The digital economy is thus run on a river of copies. Unlike the mass-produced reproductions of the machine age, these copies are not just cheap, they are free. [SNIP]

Bauwens: The result is that peer projects are sustainable collectively, but not 'individually'. As long as the projects can mobilize new volunteers to replace those they loose, the project is sustainable. However, individually of course, it is not sustainable.

Yet society is getting so many benefits from it, innovation is becoming diffuse and social and to a large extent related to precisely these peer to peer dynamics of freely contributed collaborative work, that it is worth thinking about a general basic income that would sustain it.

In the Middle Ages, this was done for about a quarter of the male population, the monks, whose 'spiritual productivity' was recognized as extremely valuable, and therefore subsidized by the whole of society. Similarly, as peer-produced social innovation is becoming more and more valuable, it will become and needs to become a process that needs to be sustained through a general investment, much like the telecommunications and transport infrastructure. I'm calling this "partner state policies which empower and enable the direct production of value".

Google and Enterprise Utility Computing p. 42

The company's latest enterprise product, Google Apps Team Edition, is a version of Google Apps that's designed for corporate users but can be used without the IT department's knowledge.

Cecil: Viewed from a larger perspective, what I thought was interesting was walled garden (corporate IT) vs the "outside". That's not to say we don't need corporate IT - I couldn't get things done without them. On the other hand, if the migration is toward cloud computing where I can pretty much do it myself with a centrally aggregated entity who can bring greater economies of scope and sale to the enterprise, then it sort of mirrors things we've discussed at the physical layer - i.e. that technology empowers end users who pretty much want big pipes and the ability to interact with whomever without intermediaries telling them how to do it.


For the complete issue you must subscribe.



The Rise of Utility Computing in the Context of the Global Grid

Could it Ever Offer Carriers an Opportunity to Cut the Umbilical to Voice?

First Some Background p. 1
Connecting the Dots or Why JP Rangaswami is SmilingTonight p. 3
Utility Based IT Services to Global Enterprises – the New Carrier Business Model? p. 4
Meanwhile Back at Home p. 5

The Network Becomes the Computer: Grid, Cloud, and Utility Computing

An Interview with 3Tera Traces the Origins Functions and Benefits of an OS for Utility Computing

Grid, Cloud and Utility Computing p. 6
Cloud Computing? p. 8
Difference Between Amazon EC and 3Tera p. 9
Genesis of 3Tera p. 10
From Scaling Online Applications to a Grid
Operating System p. 11
The Infrastructure p. 13
The Business Model p. 14

In the European Union - Regulation Will Persist in Order to Guarantee Open Access p. 18

Does the Credit Crunch Exacerbate American and Europen Tendency Toward High Return Short Term Investment?

Monetary Resource Allocation in the Credit Crunch p. 21
Consequently high return high risk is the name of the game p. 21
Business Case Complexity p. 23

Wireless: Seeing Some Good Deployment in 3.65 GHz p.29

The Economics of Triple Play, Bundles, and other Marketing Ploys p. 30

Eight Generatives Better Than Free: Is there an IP Economy Lesson here?

Eight Generatives Better Than Free p. 36

Google and Enterprise Utility Computing Sneaky or Empowering? p.42

Executive Summary p. 58

Symposium & Interview Contributors to this Issue

Affiliation given for purposes of identification - views expressed are those of the contributors alone


Burt Armijo VP Product Development 3Tera
Michel Blauwens, Chiang Mai, Thailand, Foundation for P2P Alternatives
Robert Berger, Principal & Strategist, Internet Bandwidth Development
Art Brodsky, Communications Director, Public Knowledge
Erik Cecil, Telecom Regulatory Attorney
Vint Cerf, Internet Evangelist, Google
Roland Cole, Director of Technology Policy at Sagamore Institute for Policy Research
Frank Coluccio, President, DTI Consulting Inc. South Street Seaport, NY NY
Mark Cooper, Director of Research of the Consumer Federation of America
Tim Cowen, General Counsel for BT Global Services
Robert Cromwell, Attorney City of Seattle
Vincent Dekker, teelcom report , Trouw, The Netherlands
Craig Dobson, Taylor Warwick Consulting, Alberta Canada
Harold Feld, VP media Access, Washington DC
Benoit Felten, Analyst yankee group and Fiber Revolution Blog
Fred Goldstein, Principal of Ionary Consulting, author of The Great Telecom Meltdown
Sebastian Hassinger, Stratgeist, Oracle
Conal Henry, CEO E-Net an Irish open access regional broadband infrastructure provider
Lars Hedberg, Director Swedish association of Fiber Communities
Barry Lynn, CEO 3Tera (utility computing)
Kevin Marks, Google, Coauthor Micro Formats, Quick Time Developer
Andrew Odlyzko, Director Digital Technology Center, University of Minnesota
Adam Peake, Senior Analyst Glocom, Japan
Ed Pimentel, CTO AgileCo, Alpharetta Georgia
Hendrik Rood is a senior consultant at Stratix Consulting, The Netherlands
Bill St Arnaud, Director Ca*Net4, Canada's high speed research network
John St Julien, fiber advocate and activist, Lafayette, Louisiana
Chris Savage, Partner Davis Wright Tremaine, LLP Washington, DC
Brough Turner, Co Founder NMS Communications
Jaap van Till, principal consultant and partner Stratix Consulting, Netherlands
Esme Vos, founder of,
Dirk van der Woude, Civil Servant Amsterdam and fiber expert
John Waclawsky, Chief Software Architect, Motorola
Herman Wagter, Director Citynet, AmsterdamBurt Armijo VP Product Development 3Tera