Cloudonomics E-Book

Joe Weinman

• Herausgeber: John Wiley & Sons
• Kategorie: Fachliteratur
• Sprache: Englisch

Seitenzahl: 705

Veröffentlichungsjahr: 2012

Beschreibung

The ultimate guide to assessing and exploiting the customer value and revenue potential of the Cloud A new business model is sweeping the world--the Cloud. And, as with any new technology, there is a great deal of fear, uncertainty, and doubt surrounding cloud computing. Cloudonomics radically upends the conventional wisdom, clearly explains the underlying principles and illustrates through understandable examples how Cloud computing can create compelling value--whether you are a customer, a provider, a strategist, or an investor. Cloudonomics covers everything you need to consider for the delivery of business solutions, opportunities, and customer satisfaction through the Cloud, so you can understand it--and put it to work for your business. Cloudonomics also delivers insight into when to avoid the cloud, and why. * Quantifies how customers, users, and cloud providers can collaborate to create win-wins * Reveals how to use the Laws of Cloudonomics to define strategy and guide implementation * Explains the probable evolution of cloud businesses and ecosystems * Demolishes the conventional wisdom on cloud usage, IT spend, community clouds, and the enterprise-provider cloud balance Whether you're ready for it or not, Cloud computing is here to stay. Cloudonomics provides deep insights into the business value of the Cloud for executives, practitioners, and strategists in virtually any industry--not just technology executives but also those in the marketing, operations, economics, venture capital, and financial fields.

Leseprobe

Contents

Preface

Acknowledgments

Chapter 1: A Cloudy Forecast

Clouds Everywhere

Cashing In on the Cloud

Beyond Business

Clarifying the Cloud

Farther On

Summary

Notes

Chapter 2: Does the Cloud Matter?

Productivity Paradox

Competitiveness Confrontation

Summary

Notes

Chapter 3: Cloud Strategy

Insanity or Inevitability?

Democratization of IT

Industrialization of IT

Strategy

The Cloud: More than IT

The Networked Organization

Form Follows Function, IT Follows Form

Aligning Cloud with Strategy

Everyware, Anywhere

Summary

Notes

Chapter 4: Challenging Convention

What is the Cloud?

Economies of Scale

Competitive Advantage and Customer Value

Cloud Ecosystem Dynamics

IT Spend

Issues with the Cloud

Summary

Notes

Chapter 5: What Is a Cloud?

Defining the Cloud

On-Demand Resources

Utility Pricing

Common Infrastructure

Location Independence

Online Accessibility

Difference from Traditional Purchase and Ownership

Cloud Criteria and Implications

Is the Cloud New or a New Buzzword?

Summary

Notes

Chapter 6: Strategy and Value

Access to Competencies

Availability

Capacity

Comparative Advantage and Core versus Context

Unit Cost

Delivered Cost

Total Solution Cost

Opportunity Cost and Cost Avoidance

Agility

Time Compression

Margin Expansion

Customer and User Experience and Loyalty

Employee Satisfaction

Revenue Growth

Community and Sustainability

Risk Reduction

Competitive Vitality and Survival

Summary

Notes

Chapter 7: When—and When Not—to Use the Cloud

Use Cases for the Cloud

Inappropriate Cloud Use Cases

Summary

Notes

Chapter 8: Demand Dilemma

A Diversity of Demands

Examples of Variability

Chase Demand or Shape it?

Summary

Notes

Chapter 9: Capacity Conundrum

Service Quality Impacts

Fixed Capacity versus Variable Demand

Splitting the Difference

Better Safe than Sorry

Capacity Inertia

Summary

Notes

Chapter 10: Significance of Scale

Is the Cloud Like Electricity?

Distributed Power Generation

Is the Cloud Like Rental Cars?

Capital Expenditures versus Operating Expenses

Benchmark Data

Cost Factors

Benchmarking the Leaders

Size Matters

Summary

Notes

Chapter 11: More Is Less

Is the Cloud Less Expensive?

Characterizing Relative Costs and Workload Variability

When Clouds Cost Less or the Same

If Clouds are More Expensive

Beauty of Hybrids

Cost of the Network

Summary

Notes

Chapter 12: Hybrids

Users, Enterprise, and Cloud

Hybrid Architecture Implementations

Summary

Notes

Chapter 13: Fallibility of Forecasting

Even Stranger than Strange

Demand for Products and Services

System Dynamics

Whips and Chains

Exogenous Uncertainty

Behavioral Cloudonomics of Forecasting

Summary

Notes

Chapter 14: Money Value of Time

Demand and Resource Functions

Cost of Excess Capacity

Cost of Insufficient Capacity

Asymmetric Penalty Functions, Perfect Capacity, and On Demand

Flat Demand

Uniformly Distributed Demand

Better Never than Late

MAD about Being Normal

Triangular Distributions

Linear Growth

Exponential Growth

Random Walks

Variable Penalty Functions

Summary

Notes

Chapter 15: Peak Performance

Relationships between Demands

Lessons from Rolling Dice

Coefficient of Variation and Other Statistics

Statistical Effects in Independent Demand Aggregation

Significance of

Issues with Perfectly Correlated Demand

Community Clouds

Simultaneous Peaks

Peak of the Sum is Never Greater than the Sum of the Peaks

Utilization Improvements

Summary

Notes

Chapter 16: Million-Dollar Microsecond

On Time

Rapidity Drives Revenue

Built for Speed

Summary

Notes

Chapter 17: Parallel Universe

Limits to Speedup

Amdahl versus Google

Free Time

Summary

Notes

Chapter 18: Shortcuts to Success

Rapid Transit

Sending Letters

Short on Time

Bandwidth Isn’t Enough

Summary

Notes

Chapter 19: Location, Location, Location

Latency and Distance

Circle Covering and Circle Packing

Inverse Square Root Law

Spherical Caps and the Tammes Problem

Summary

Notes

Chapter 20: Dispersion Dilemma

Strategies for Response Time Reduction

Consolidation versus Dispersion

Trade-offs between Consolidation and Dispersion

Benefits of Consolidation

Benefits of Dispersion

The Network Is the Computer

Summary

Notes

Chapter 21: Platform and Software Services

Infrastructure as a Service Benefit

Paying on Actuals versus Forecasts

Installation

Investment

Updates

Service-Level Agreements

Continuously Earned Trust

Visibility and Transparency

Big Data and Computing Power

Ubiquitous Access

Response Time and Availability

Multitenancy, Shared Data

Cloud-Centric Applications

Scalability

Communities and Markets

Lock-in

Security and Compliance

PaaS: Assembly versus Fabrication

Innovation and Democratization

Deconstructing the Pure SaaS Model

Summary

Notes

Chapter 22: Availability

Uptime versus Downtime

Availability and Probability

Availability of Networked Resources

Availability via Redundancy and Diversity

On-Demand, Pay-per-Use Redundancy

Summary

Notes

Chapter 23: Lazy, Hazy, Crazy

Behavioral Economics

Loss and Risk Aversion

Flat-Rate Bias

Framing and Context

Need for Control and Autonomy

Fear of Change

Herding and Conformity

Endowment Effect

Need for Status

Paralysis by Analysis of Choice

Hyperbolic Discounts and Instant Gratification

Zero-Price Effect

Summary

Notes

Chapter 24: Cloud Patterns

Communications Patterns

Hierarchies

Markets

Repository

Perimeters and Checkpoints

Summary

Notes

Chapter 25: What’s Next for Cloud?

Pricing

Ecosystems, Intermediaries, and the Intercloud

Products versus Services

Consolidation and Concentration

City in the Clouds

Spending More while Paying Less

Enabling Vendor Strategies

Standards, APIs, Certification, and Rating Agencies

Commoditization or Innovation?

Notes

About the Author

About the Web Site

Index

Additional praise for Cloudonomics: The Business Value of Cloud Computing

It is a business imperative to do more with less—and do everything faster. Cloudonomics offers a much-appreciated framework for sorting through cloud computing options and the marketing hype.

—Lorraine Cichowski, SVP and CIO, Associated Press

Joe Weinman is one of the foremost thinkers in cloud computing. He has captured a topic of fierce complexity and expressed it with elegant prose and simple, powerful, and compelling mathematical rigor. Weinman delivers his arguments with a clarity and logic that is unassailable. Cloudonomics should be required reading for every enterprise CIO seeking a way through the clutter and hype of vendors’ cloud solutions and looking for a set of crystal clear, mathematically grounded, and meticulously presented arguments that show the way through the morass of cloud computing.

—Simon Crosby, PhD, CTO, Bromium; founder and former CTO, XenSource; former CTO, Data Center and Virtualization, Citrix

Cloudonomics is a seminal work on cloud based on an axiomatic mathematical theory, and not on popular opinions and baseless assumptions. Based on rigorous quantitative analysis, the book is amazingly simple to read with real-world examples in a lucid language that will resonate with both the technical and business professional. This book is a must-read for every professional interested in cloud and is an invaluable reference for any advanced course on cloud computing.

—Ravi Rajagopal, VP, Cloud Strategy and Solutions, CA Technologies; Adjunct Professor, New York University

Joe Weinman’s masterful book looks past the hype to offer new insights into the impact of cloud computing. Cloudonomics is must-reading for anyone interested in a more analytically based understanding of the cloud’s transformative potential.

—Christopher S. Yoo, John H. Chestnut Professor of Law, Communication, and Computer & Information Science and founding director of the Center of Technology, Innovation and Competition, University of Pennsylvania

The cloud is redefining how technology is being used for businesses. Cloudonomics is a book that will help you understand the economics and importance of this change and what it means for your industry.

—Om Malik, founder, GigaOM & Structure: the Cloud Computing conference

Copyright © 2012 by Joe Weinman. All rights reserved.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey.

Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, [phone], fax [phone], or on the Web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, [phone], fax [phone], or online at http://www.wiley.com/go/permissions.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at [phone], outside the United States at [phone] or fax [phone].

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. For more information about Wiley products, visit our web site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Weinman, Joe, -

Cloudonomics: the business value of cloud computing / Joe Weinman.

p. cm.

Includes index.

ISBN 978-1-118-22996-5 (pbk); 978-1-118-28696-8 (ebk); 978-1-118-28402-5 (ebk); 978-1-118-28288-5 (ebk);

1. Cloud computing—Economic aspects. 2. Information technology—Management. I. Title.

QA76.585.W45 2012

004.6782—dc23

2012012399

To Paige, Ali, and Sierra

Preface

In the course of human history, there have been a number of bona fide revolutions in the interdependent arenas of technology, society, religion, economics, and politics: flint tools, money, writing, agriculture, democracy, printing, steam power, capitalism, mass production, telephony, and electricity, to name a few. We are 65 or 70 years into one such revolution—the information age—which has permeated every corner of the earth and beyond—from video games to war games to baseball games, and from first-world stock markets to third-world fish markets, to out-of-this-world interstellar probes.1 Oh yes, and musical greeting cards, talking dolls, and intelligent thermostats too.2

Does the advanced age of this advanced age signal impending retirement? Some argue that “the opportunities for gaining IT-based advantages are already dwindling,”3; however, this sounds suspiciously similar to alleged pronouncements, such as “everything that can be invented has been invented” or “there is a world market for maybe five computers.”4

The revolution is accelerating, not slowing.

Technologies such as quantum computing, digital electro-holographic displays, brain-computer interfaces, natural-language interaction via speech-to-text and semantic processing, homomorphic encryption, and new electronic components such as HP’s nanoscale memristor, Intel’s three-dimensional chips, and on-chip optical interconnects are still in their infancy. Innovative cognitive computers are now being designed by IBM to use “brainlike” neurosynaptic chips. Other disturbing anomalies, from quantum entanglement to apparently faster-than-light neutrinos, may form the foundation for future disruption.5

In this ocean of innovation, cloud computing is the latest of successive waves that have eroded the shoreline of prior paradigms, such as the mainframe, the minicomputer, and the personal computer.

Cloud computing is a tsunami of transformation exemplifying Schumpeterian creative destruction: amassing immense wealth for companies that didn’t exist a few years ago—such as Google, Facebook, Amazon, Salesforce.com, and Zynga; disrupting long-standing business models and ecosystems including publishing, advertising, television, the recording industry, telecommunications, and retailing; and reordering relationships within the computing industry: among hardware vendors, licensed software vendors, distributors, value-added resellers, and systems integrators, to name a few.

The cloud is both an existential threat and an irresistible opportunity. Virtually any summary of key trends or chief information officer (CIO) focus areas ranks cloud computing at or near the top of the list. A recent Gartner survey of 2,000 CIOs places cloud computing as the number-one technology priority.6 Most if not all of the rest of the top priorities—virtualization, mobility, collaboration, business intelligence—enable, are enabled by, or otherwise relate to the cloud.

Wharton fellow and author Jeremy Rifkin would consider this to be a natural consequence of “the Age of Access.”7 He has argued that the market economy—in which people own and trade goods—is being replaced by the network economy—where people pay to access them. Why bother owning something if you can access it anytime, anywhere? People don’t want drills but the holes that they make; people don’t want CDs or applications but the music or functionality that they provide.

Although Rifkin has positioned this trend as something new, in many respects it represents a return to a prior age. After all, the idea of content ownership is relatively new: Before owning records, CDs, or MP3s, people accessed audio content via the concert hall or radio; before VHS, Beta, or DVDs, there was broadcast and cable television, movie theaters, and, even earlier, plays and operas. In the Age of Access 2.0, however, the logic, characteristics, and payment models are certainly changing. Rather than traveling to theaters and opera halls, the content comes to you; unlike broadcast, it is personalized, contextualized, and on demand.

So the cloud is the new, new thing, but what it actually is, is a matter of disagreement. Cloud computing, so named by Ram Chellappa of Emory University, is, at a high level, computing that is done somewhere out there in an undisclosed location away from your own laptop, smartphone, tablet, or data center.8 The cloud model applies to the discovery and acquisition of applications, services, and content, such as eBooks from ebookstores; tablet and smartphone apps from app stores; songs from Lady Gaga on Vevo or blockbuster movies on demand; and customer relationship management software executed far from your device. The cloud is at the heart of social networks such as Facebook and LinkedIn, social games such as Zynga’s FarmVille, microblogs such as Twitter, and texting, messaging, and mail such as AOL Instant Messenger, RIM’s BlackBerry Messenger, Microsoft Hotmail, and Google Gmail. But it also applies to the core infrastructure (computer servers and data storage), the utility software (middleware and databases), and the currency of the digital economy, “big data” (petabytes of information), that together enable those applications to run at the scale of millions or hundreds of millions of users.

Like the parable of the blind men describing an elephant—one feels the legs and says the elephant is like a pillar, the other the tail and says it is like a rope—or perhaps like real cloud gazers—one sees a rabbit, one his Aunt Martha—everyone has a different perspective on the cloud. Some see a new technology, say, virtualization; others, a new silo-busting integrated development and operations model; others, a means of delivering software functionality; still others, an ecosystem that enables applications spanning mobile devices, networks, and cloud-based resources and services. They are all correct in their own way.

For the purposes of this book, however, we consider cloud computing primarily from the business, financial, and economic perspective: Cloudonomics, to use the term I coined in the summer of 2008 for Cloudonomics.com and a blog post for the popular technology site GigaOM.com, syndicated to BusinessWeek.9 As such, we consider core characteristics of the cloud—on-demand resources, usage-based charging, resource sharing, geographic dispersion, and the like—and how they map to and drive business—and even societal—value.

I would claim that such a perspective is one of the most important ways to evaluate and exploit the cloud, since unless a technology drives compelling value, it will end up in the dustbin of history. Remember the CueCat barcode reader?10

The value of the core characteristics of the cloud has been proven time and again in domain after domain: hotel chains, airlines, electric utilities, taxi services, manufacturing service providers, and others. Taxis offer transportation capacity on an on-demand, pay-per-use basis. Banks rent resources as well—principal—on a pay-per-use basis—interest. Companies can buy workers’ services from what could be called the “labor cloud” either on a flat-rate basis—a salary—or a pay-per-use basis—hourly wages.

Ubiquitous access and location independence are key. When you order a physical book from Amazon.com or DVD from Netflix, you don’t really need to know what distribution center it was sent from as long as it arrives on time, and when you order an eBook from Amazon.com or stream a video from Netflix, you don’t really need to know what data center it came from. Either way, you trust that the provider has figured out the appropriate locations to get you what you want within the time frame that you would like and to which you agree via the terms of service.

In the pay-per-use model of the cloud, we see the same charging model used by hotels and barber shops. In on-demand provisioning, we see the same resource allocation strategy used in accessing energy resources by turning on an electric switch, or financial resources by tapping into a home equity line of credit. Geographically dispersed data centers and content-delivery networks echo the approach used by coffee shops and fast food chains to distribute their wares globally. Resource sharing of computer servers in a cloud data center is not that different from sharing servers—waiters and waitresses—in a restaurant.

These are more than casual analogies; the point is that the same immutable principles—say, resource utilization improvements from demand aggregation or diminishing returns from investments in geographic dispersion for latency reduction—apply regardless of domain. I call these the Laws of Cloudonomics. The Laws of Cloudonomics are not restricted to cloud computing any more than the Law of Gravity is restricted to apples.

From these parallels—and an analysis of underlying, abstract models—we can determine that there are quite a few characteristics and behaviors that contravene simplistic thinking, a sort of freakonomics of the cloud. For example, branch expansion is doomed to fail eventually. Rational customers often should be delighted to pay more for cloud services—really—even when there are no differences in characteristics such as performance or security. It can cost nothing to accelerate performance. Even after both heavy and light users switch rate plans from flat rate to pay-per-use or vice versa to save money, a firm or industry can maintain revenues. As the cost of IT plummets, IT spend will stay the same—or increase.

This book doesn’t focus on industry market projections or vendor offerings but rather on strategy, business models, customer value, and their relationships. The intent of this book is to be multidisciplinary, seminal, evergreen, rigorous, forward-looking, and irreverent and to appeal to a broad range of customers, prospects, strategists, venture capitalists, investors, technologists, executives, service providers, and academics, both within the field of cloud computing and beyond.

This book is first and foremost multidisciplinary, drawing on illustrative industry examples but also a broad range of fields: strategy, economics, psychology, system dynamics, calculus, statistics, computing technology, and theoretical computer science, with forays into botany, biology, and physics. Most readers should find most of the book quite readable, and experts in a variety of fields should find the breadth of Cloudonomics of interest.

Second, I hope that much of this book is seminal. I believe I’ve been first to explore a number of areas as they relate to the cloud, such as the architectural implications of cost optimization, analysis of latency for interactive applications using packing of spherical caps, and computational intractability of networked resource allocation.

Third, rather than conducting a Consumer Reports–style analysis of ratings of the different vendors and service providers that would be out-of-date before the book is even published, the intent is for the book to be evergreen (i.e., usable for a long time to come) in assessing architectural and business alternatives, developing new business models, and incorporating cloud into your own business strategy.

The suffix “-nomics” has been used for important insights into business trends and strategy—Eric Qualman’s Socialnomics on social media, Ken Doctor’s Newsonomics regarding the digitalization of news, and Don Tapscott and Anthony Williams’ Wikinomics covering IT-based collaboration—to the more quantitative Freakonomics, presidential economic policies such as Reaganomics, Clintonomics, and Obamanomics, and humorous insurance advertising—Geckonomics. However, Cloudonomics is quantitatively rigorous. I’ve skimmed the surface of much of the math for this volume. However, it is worth noting that most claims are supported by detailed proofs. My research has taken me far afield of “the cloud” and uncovered unusual connections. For example, cell site or content delivery network node placement is related to cannonball stacking, the Tammes problem—solved by evolution—of designing pollen spores, and the Thomson problem of minimum energy electron configurations, which in turn relates to baryon density isosurfaces of Skyrmions, whatever they are.11

References are provided to more detailed papers and to my simulation Web site: ComplexModels.com, which provides easy-to-use Monte Carlo simulations illustrating the Laws of Cloudonomics.

Of course, being overly quantitative can lead to precise yet incorrect conclusions due to quirks, biases, and anomalies in human behavior, which I will touch on. These experiments have led to more than one Nobel Prize in economics12 and provide fascinating insights into what Duke University professor Dan Ariely calls “predictable irrationality.”13

Cloudonomics posits forward-looking scenarios for cloud computing and industry ecosystem evolution and revolution. For example, braking at high speed through a turn on a wet road used to be the domain of professional stunt drivers; now any driver can perform at the same level using standard antilock braking systems (ABS), electronic stability control, and traction control. Similarly, proprietary virtual server provisioning tools, for example, are not likely to offer sustainable competitive advantage to a cloud service provider. This technology is diffusing just like ABS.

Finally, the book is irreverent, challenging conventional wisdom. The truth in cloud is often counterintuitive and nuanced; better mental models mean better business and technology strategies and investments.

I hope to provide a lens to view the world of cloud computing economics, so that you can consider my questions and evaluate my arguments but draw your own conclusions. I do not necessarily provide all the answers but present models with which you can perceive, comprehend, and thereby exploit the cloud for your own uses, whether as a customer, service provider, equipment manufacturer, software vendor, venture capitalist, or investor. This is not the end of the dialogue, but the beginning.

1Michael Lewis, Moneyball: The Art of Winning an Unfair Game (Norton, 2003). Robert Jensen, “The Digital Provide: Information (Technology), Market Performance, and Welfare in the South Indian Fisheries Sector,” Quarterly Journal of Economics 122, No. (2007): 879–924.

2www.nest.com/living-with-nest/.

3Nicholas Carr, “IT Doesn’t Matter,” Harvard Business Review (May 2003). 81(5) 41–49.

4These quotes are of questionable authenticity. See Kevin Maney, “Tech Titans Wish We Wouldn’t Quote Them on This Baloney,” USA Today, July 5, 2005. www.usatoday.com/tech/columnist/kevinmaney/2005-07-05-famous-quotes_x.htm.

5Adrian Cho, “From Geneva to Italy Faster than a Speeding Photon?” Science, September 30, 2011, Vol. 333, No. 6051, p. 1809.

6Gartner, “Gartner Executive Programs Worldwide Survey of More than 2,000 CIOs Identifies Cloud Computing as Top Technology Priority for CIOs in 2011,” Gartner Newsroom, January 21, 2011. http://www.gartner.com/it/page.jsp?id=1526414.

7Jeremy Rifkin, The Age of Access: The New Culture of Hypercapitalism, Where All of Life Is a Paid-For Experience (Tarcher, 2001).

8www.bus.emory.edu/ram/.

9Joe Weinman, “The 10 Laws of Cloudonomics,” GigaOM.com, September 7, 2008, http://gigaom.com/2008/09/07/the-10-laws-of-cloudonomics/. Joe Weinman, “The 10 Laws of Cloudonomics,” BusinessWeek, September 6, 2008, www.businessweek.com/technology/content/sep2008/tc2008095_942690.htm.

10Dan Tynan, “The 25 Worst Tech Products of All Time,” PCWorld, May 26, 2006. www.pcworld.com/article/125772-8/the_25_worst_tech_products_of_all_time.html.

11Gerald Edward Brown and Mannque Rho, The Multifaceted Skyrmion (World Scientific Publishing, 2010).

12Technically, the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, nobelprize.org.

13Dan Ariely, Predictably Irrational: The Hidden Forces that Shape Our Decisions (HarperCollins, 2008).

Acknowledgments

A book like this owes so much to so many that it is impossible to fully trace the directed causal graph. First and foremost, however, I’d like to thank the wonderful team at John Wiley & Sons, Inc. Sheck Cho, executive editor at Wiley, immediately saw the potential of this book and has been extremely professional, flexible, insightful, transparent, collaborative, and patient. I’d also like to thank the rest of the terrific Wiley team, especially Natasha Andrews-Noel, Stacey Rivera, and Helen Cho, who helped make this book a reality. Thanks to Bennett Ruiz of AT&T, Barrie Sosinsky, and Hunter Muller of HMG Strategy for helping make the connection with Wiley. And thanks to Zick Rubin and Brenda Ulrich at the Law Office of Zick Rubin, who were both knowledgeable and responsive.

I believe that a book such as this is immeasurably enriched by data. Beyond the extensive references, thanks are particularly due Greg Orelind of Alexa, who kindly permitted the use of the Alexa pageview data illustrating demand variability; Marty Kagan and Greg Unrein of Cedexis for HTTP response time data; Ali Kafel and Dave Connolly of Sonus Networks; Stephan Beckert and Olivia Vandenbussche of TeleGeography; and James Miller of the FCC.

The usual disclaimers apply; I take full responsibility for any errors, which, sadly, have a nonzero probability of existing in a book of this scope.

Any delineation of the main causal path of events leading to my involvement in the cloud would have to include Eric Shepcaro, Allan Leinwand, Om Malik, and Alistair Croll. I worked for Eric beginning at the turn of the millennium when he was AT&T’s senior vice president of application networking. We were pioneers in introducing new hosting services—called utility computing at the time—and, thanks to Eric, I was an active participant in Don Tapscott’s IT & Competitive Advantage program—the syndicated research effort that led to Wikinomics and included in-depth collaboration with a host of thought leaders: Don himself, David Ticoll, Joe Pine, Anthony Williams, Rena Granofsky, Paul Strassmann, Erik Brynjolfsson, Charlie Fine, Mike Dover, and others.

Eric also introduced me to Chris Albinson and Allan Leinwand of Panorama Capital, who invited me to join their Technology Advisory Board. Allan also introduced me to Om Malik. My first official cloud event was Om’s Structure, in June 2008, where I was on a panel moderated by Alistair, who asked a number of thought-provoking questions, which in turn led to my first blog posts for GigaOM.com, including “The 10 Laws of Cloudonomics.” This book is a 100,000-word-plus expansion of the “Laws” post and a number of others I’ve done for GigaOM.com.

Om is the epicenter of the cloud, between his own social and professional network, the focus GigaOM.com and GigaOM Pro put on it, and the Structure event, which I’ve had the pleasure of participating in as MC, moderator, and panelist since its inception. At Giga Omni Media, I’ve had the good fortune to work with Paul Walborsky, Stacey Higginbotham, Surj Patel, Derrick Harris, Carolyn Pritchard, Celeste LeCompte, Mike Sly, a host of cloud innovators and executives, and the magnificent Magnify Communications team: Stacey Tomlinson, Erin McMahon Lyman, and Jill Short Milne.

Extra-special thanks to Carl Brooks (now at The 451 Group) and Jo Maitland (now at GigaOM) of TechTarget, who were overly kind in naming me to their prestigious list of Cloud Computing Leaders.

I’d also like to thank UBM TechWeb, which owns and operates Interop, Cloud Connect, Light Reading, Heavy Reading, InformationWeek, and the former Business Communications Review (BCR, now moved online as No Jitter). Particular thanks go to Alistair Croll of Bitcurrent, Eric Krapf of No Jitter, and Blair Klein of AT&T. Alistair is always thinking several steps ahead of everyone else, and has been organizing both the Interop cloud summits as well as TechWeb’s Cloud Connect, where I’ve served as track chair of the cloud economics track and had the opportunity to learn from many cloud thought leaders and to work with the TechWeb events leadership and crew: Lenny Heymann, Steve Wylie, Manuela Farrell, Andy Saldana, Paige Finkelman, Amy Jones, and Emily Johnson. Blair is AT&T’s social media expert and also PR lead; it was she who originally asked me “Have you ever considered writing anything for external publication?” and suggested BCR; Eric Krapf was then editor in chief and published my first cloud article—“The Evolution of Networked Computing Utilities”—before cloud was cool. At Light Reading, special thanks to Ray Le Maistre and Carol Wilson, at InformationWeek, John Foley and Charlie Babcock.

At AT&T Labs, I was also fortunate to work with Ralph Wyndrum, Norm Shaer, Rick Kent, Rudy Alexander, Bernie McElroy, Stan Quintana, Steve Fisher, Sam Glazer, Clayton Lockhart, Tom Siracusa, Dave Belanger, Chuck Kalmanek, Ed Amoroso, Rick Schlichting, and dozens of other technical experts and executives and to interact with literally thousands of AT&T customer executives and their teams—in groups ranging from 1 to 1,000—thanks to executive/sales senior leaders Ron Spears, Donna Henderson, John Finnegan, Bennett Ruiz, Andrea Messineo, Norihiko Minato, Gopi Gopinath, Bernard Yee, and Andrew Edison; countless excellent country managers; regional sales vice presidents; sales teams; and PR, marketing, events, and legal leads Greg Brutus, Peter J. Butler, June Chan, Fenny Fang, Karen Ko, Mary Beth Asher, Dagmar Hettler-Gentil, Niall Hickey, Donna Cobb, Eileen Whelan, Andrea Montesano, Christine O’Leary, Linda Plesnick, Don Parente, Joanne Murphy, Wendy Weinstein, Janet Wyles, Sara Vincent, and Karen Wentworth. I enjoyed working with AT&T strategy, hosting, and cloud product management executives Bill Archer, Steve Sobolevitch, Steve Mucchetti, Steve Caniano, Chris Costello, Tim Connors, Amy Anderson, Toby Ford, and others. I was fortunate to work with the team at Fleishman Hillard, including Morri Berman, Patrick Yu, Gioconda Beekman, Winnie Leung, and Brad Mays.

At HP, I had the pleasure of working with great colleagues: Erwan Menard, Sandeep Johri, Dave Shirk, Rick Halton, Russ Daniels, Dave Collins, Sybille Schieg-Heimann, Reem El-Tonsy, Julia Ochinero, Andrea Nicole Garcia, Paul Battaglia, Blaithin Underhill, Rebecca Lawson, and Emil Sayegh, as well as my global team and business and technical colleagues. Special thanks to Jujhar Singh for his support in permitting me to work on the book in my spare time and to Jan Tarantino for her critical role. As at AT&T, at HP I also had the pleasure of working with a large group of excellent regional sales VPs, country managers, account teams, and cloud experts too numerous to name. And I also had the superb support of IVORY Europe: Andrea Lee, Kristina Dalborg, Robbie Crittall, and Harry Whitbread.

I’ve been involved in dozens of cloud initiatives, events, planning committees, and conferences that have helped introduce me to customers with unique problems and insights and cloud thought leaders and innovators. In addition to GigaOM Structure and Techweb/UBM events, I’ve been fortunate to participate in numerous events and initiatives. Some highlights: Debbie Landa and her team at Dealmaker Media/Under the Radar; George Gilder and Telecosm; Sharon Nakama, Gary Kim, Anamarcia Lacayo, and the rest of the team at the Pacific Telecommunications Council; Frank Gens at IDC and John Gallant at IDG Enterprise; Karen Tucker and the Churchill Club; Stuart Sharrock and Katz Kiely of ICIN and ITU; Tier 1/The 451 Group; CDM Media; TM Forum; IIR/Informa; Cloud Expo; Milken Global Institute; Mobile World Congress; BSS/OSS World; SIIA/All About the Cloud; GDS International; Argyle Executive Forum; Simon Torrance and STL Partners/Telco 2.0; Capacity Media; SWIFT/Sibos, and IEEE Technology Time Machine conference planners including Rico Malvar, Gerhard Fettweis, Maurizio Decina, and Roberto DeMarco.

I’ve also learned immensely from the academic/technical community: Special thanks to Ravi Rajagopal, a VP at CA Technologies but also an adjunct professor at NYU Polytechnic and the first to use my Cloudonomics papers in academia; Christopher Yoo, polymath professor of law at the University of Pennsylvania; Michael Lightner, past president of the IEEE and a professor at University of Colorado, Boulder, who gave selflessly of his time in ensuring the rigor and breadth of axiomatic cloud theory; Rico Malvar of Microsoft Research and the IEEE Technology Time Machine team including faculty of the Hong Kong University of Science and Technology; and Lucy Hood of the University of Southern California Marshall School.

The home of cloud computing, via both the invitation-only @Clouderati group and as an extended community is primarily on Twitter, where I am @joeweinman. Most of these individuals are also bloggers, executives, research analysts, CTOs, CIOs, and consultants. A special glue holds this group together, and I consider these leading cloud experts to be not only colleagues but friends and thought-leaders. Literally from A to Z, they are: Vanessa Alvarez, Dave Asprey, David Berlind, Linda Bernardi, Randy Bias, Benjamin Black, Simone Brunozzi, Rachel Chalmers, Sam Charrington, Jean-Luc Chatelain, Adrian Cockcroft, Peter Coffee, Reuven Cohen, Tim Crawford, Simon Crosby, Ellen Daley, William Fellows, Rodrigo Flores, Will Forrest, Mike Fratto, Jay Fry, Vijay Gill, Barton George, Bernard Golden, Charles Golvin, Christofer Hoff, Jason Hoffman, Sam Johnston, Ton Kalker, Jeff Kaplan, John Keagy, Ben Kepes, Michael Krigsman, Maribel Lopez, Bob Lozano, Tom Lounibos, William Louth, Barry Lynn, Lori Mac Vittie, Dave McCrory, Dave McCandless, Brenda Michelson, Marten Mickos, Rich Miller (Data Center Knowledge), Rich Miller (Telematica), Lew Moorman, Ofir Nachmani, Greg Ness, Vinay Nichani, Deb Osswald, Geva Perry, Antonio Piraino, Justin Pirie, Eric Pulier, Elizabeth Rainge, Surendra Reddy, Christian Reilly, Dave Roberts, Guy Rosen, Scott Sanchez, Steve Shah, James Staten, Krishnan Subramanian, Shlomo Swidler, Mark Thiele, Lew Tucker, James Urquhart, Jinesh Varia, Werner Vogels, Simon Wardley, James Watters, John Willis, Dan Young, Jian Zhen, Joe Ziskin.

Thanks to the cloud itself for helping research and connect, not just Twitter, but Google, Google Books, Google Scholar, Alexa, JSTOR, Wikipedia, and hundreds of other online resources such as newspaper and magazine articles.

Final thanks go to my family. To my wife, Carolyn, for helpful editorial comments and for her patience as I spent every night, weekend, and vacation day for six months typing away from behind a wall of books, in the presence of a monotonically increasing list of delayed household projects. To my mother, Elisa, and late father, Joe, for being two wonderful, loving, dedicated parents. Finally, to my daughters, Paige, Ali, and Sierra, for their love and support.

About the cover: The cover illustrates the three perspectives of the book: the real world, shrouded in actual meteorological clouds, symbolizing objectives such as optimized customer experience and issues such as cognitive biases; the world of IT, symbolized by a circuit board; and the abstract mathematics underlying both of them, symbolized by the grid.

CHAPTER 1

A Cloudy Forecast

The cloud—shorthand for “cloud computing”1—is transforming all spheres of our world: commerce, entertainment, culture, society, education, politics, and religion. Cloud start-ups are forming on a daily basis, and billions of dollars in wealth are being created as companies craft innovative strategies to exploit this opportunity. Conversely, long-standing corporate icons that have failed to do so are becoming history instead of making it.

The concept of a public cloud—shared, on-demand, pay-per-use resources, accessible over a wide-area network, available to a broad range of customers—might appear to be a recent breakthrough, but there is nothing new under the sun, not even the cloud. The ancient Romans implemented the information superhighway of their time, constructing an unprecedented wide-area network with thousands of route miles of roads, called viae, using state-of-the-art engineering, following documented standards.2 The public network, made of public roads, or viae publicae, was complemented by and interoperable with metro networks, the viae vicinales, and private networks, the viae privatae, creating an Internet of sorts. The roads of the Romans carried people, goods, and soldiers, but, perhaps most important, they also served as a communications network, enabling information, coordination, and control of the far-flung republic and then empire.

These viae were multiprotocol networks—carrying pedestrians, animals, carts, military chariots, horses, and their riders—with class of service—military and chariots in the center lane, pedestrians and slower vehicles to the side.3 Net neutrality was assumed: Any citizen could traverse the viae publicae and even had certain rights of passage on the viae privatae.4 By order of Caesar, the core of the network had congestion management: Transport carts were banned from the network core—the narrow, winding streets at the heart of Rome—from dawn until dusk.5 A complementary architecture was used for streaming content delivery: the aqueducts.