You are on page 1of 84

Contents

Chapter 1 ................................................................................................................................................. 2
WHAT IS A CLOUD?........................................................................................................................ 2
HYPE CYCLE ................................................................................................................................ 3
IMPLEMENTATION GAP ............................................................................................................ 5
COMMON DEFINITION............................................................................................................... 6
ATTRIBUTES ................................................................................................................................ 8
RELATED TERMS ...................................................................................................................... 10
HISTORY ..................................................................................................................................... 11
CLOUD ARCHITECTURE ............................................................................................................. 14
STACK ......................................................................................................................................... 14
MANAGERMENT LAYER ......................................................................................................... 18
PRIVATE, PUBLIC, AND HYBRIDS CLOUDS ....................................................................... 19
TOPOLOGY ................................................................................................................................. 22
ECOSYSTEM ............................................................................................................................... 24
SERVICE MODELS OF CLOUD COMPUTING ........................................................................... 26
INFRASTRUCTURE AS A SERVICE ( IaaS) ............................................................................ 26
PLATFORM AS A SERVICE ( PaaS) ......................................................................................... 34
SOFTWARE AS A SERVICE ( SaaS) ......................................................................................... 37
Chapter 2 MOVING TO THE CLOUD ................................................................................................ 39
A BUSINESS STRATEGY BASED ON AGILITY ........................................................................ 39
USING THE CLOUD FOR BUSINESS ADVANTAGE ................................................................ 46
BUSINESS APPLICATIONS WITH THE GREATEST POTENTIAL .......................................... 52
CONTEMPORARY WAYS BUSINESS ARE USING CLOUD .................................................... 54
Chapter 3 GLOBAL IMPLICATIONS OF THE CLOUD ................................................................... 60
REAL- TIME GLOBAL COLLABORATION ................................................................................ 60
NEW WAY OF WORKING............................................................................................................. 64
NEW REALITIES AND NEW OPPORTUNITIES ......................................................................... 71
Chapter 4 FUTURE TRENDS .............................................................................................................. 76
ECONOMIC ..................................................................................................................................... 76
CULTURE ........................................................................................................................................ 78
VIRTUAL AND PHYSICAL CONVERGENCE ............................................................................ 79
NEXT-GENERATION NETWORKING ......................................................................................... 80
1

PEER-TO-PEER NETWORKING ................................................................................................... 83

DEFINITION
INTRODUCTION
PHAN KALI

Chapter 1
WHAT IS A CLOUD?
So, what is ―Cloud Computing‖? As we shall see, the experts disagree on its exact definition
but most concur that it includes the notion of web-based services that are available ondemand from an optimized, and highly scalable, service provider. Despite the technical sound
of its characterization, it has not only garnered excitement from technologists but has
captured the attention of business leaders around the world.
2009 may long be remembered as the year of the worldwide economic downturn. The impact
across many industries, including IT, was devastating. And yet, even then, there were
bastions of hope in the midst of all the turmoil. One of the most prominent IT analysts,
Gartner, identified cloud computing as the primary source of growth in IT spending,
increasing over 20% year-on-year to global revenues of $56B in 2009 and surging to over
$130B by 2013 .
In other words, while IT managers have seen their budgets decimated in the course of intense
scrutiny over any uncritical or non-urgent costs, they have actually received permission to
spend more in a completely new area that didn‘t even appear in any of their proposals a few
years ago.
2

HYPE CYCLE

At the Gartner Emerging Technologies conference in 2008, analysts Daryl Plummer and
Thomas Bittman made a bold declaration that: ―By 2012, 80 percent of Fortune 1000
companies will pay for some cloud computing service, and 30 percent of them will pay for
cloud computing infrastructure‖.
While it is always difficult to predict the future, the signs are appearing that cloud computing
is poised for a breakthrough. Before you bet your business on it, however, it‘s worth
investigating the relationship between the excitement around new technologies and actual
investment.

Figure 1-1 Cloud Hype Cycle

Industry analysts use a ―Hype Cycle‖ (originally developed by Gartner and loosely depicted
in Figure 1.1) to describe the tendency for new technologies to garner an excess of interest
long before they are mature enough to be deployed in production environments. The
disappointment that many advocates feel when their initial zeal is frustrated leads them into a
―Trough of Disillusionment‖. During this phase many lose heart and redirect their interest
3

toward other promising ventures. However, as the underlying problems are resolved and
aligned with user needs the systems become ready for the mainstream and interest picks up
again.
There is an element of subjectivity around the precise positioning of any technology in the
cycle. Gartner placed cloud computing in the middle of the initial slope in 2008 and advanced
the position to the peak of the hype curve in 2009. Other analysts loosely concur with this
assessment – there is little disagreement on the level of enthusiasm around cloud computing,
even if it may already be seeing some signs of the onset of disillusionment in 2010.
A more objective, albeit not necessarily extremely reliable, way to gauge the excitement
around any given topic is to use Google Trends (Figure 1.2). You can enter a search
expression and Google will provide a graphic representation that illustrates the relative
frequency of search of this specific expression over time.

Figure 1-2 Google Trend "Cloud Computing"

For instance, the figure below illustrates the popularity of the phrase ―Cloud Computing‖
over the past five years. Until the middle of 2007, the term was virtually never used in search
queries nor did it appear in news references. Since then, the search requests and appearances
in news articles have grown dramatically.

4

Early Adopters. While being at one extreme or the other may give cause for some concern there is not necessarily any value associated with being earlier or later in the adoption curve. Technology enthusiasts often embrace new developments before the commercial business case becomes compelling.IMPLEMENTATION GAP While it is very interesting to understand the hype around a new technology. Some companies can gain competitive advantage by leveraging and showcasing the newest technologies. and Late Majority. Laggards) to designate these segments. Figure 1-3 Moore's Implementation Chasm Moore segments the target market according to the speed with which they adopt new technologies. Moore‘s salient observation is that there is a significant barrier to overcome between the Early Adopters and Early Majority. Others may risk their business if they pursue areas outside their core competency too early. Early Majority. 5 . it is also helpful to examine the historical implementation cycle of new technologies. He uses five categories (Innovators. pave the way to widespread adoption which will be recognized and further refined by an Early Majority. alongside a reliable and easily deployable technological foundation. A generally applicable business case.

Everyone has a different definition of cloud computing. some definitions which have circulated in the blogosphere include: The 451 Group: ―The cloud is IT as a Service.An interesting twist that cloud computing brings to the adoption model is that. It isn‘t hard to find an authoritative definition. And still the most common question I hear when I mention the topic is ―what does it actually mean?‖ It makes sense to start this discussion with the same question. COMMON DEFINITION We have established that ―Cloud Computing‖ is one of the buzzwords for 2010. This can be attractive to those Luddites who would prefer to concentrate on non-technical aspects of their business and are glad to see their fixed IT costs reduced and their computer departments. After all. minimized. delivered by IT resources that are independent of location‖ Gartner: ―Cloud computing is a style of computing where massively scalable IT. Unfortunately the task of defining cloud computing is much more difficult than you might expect. that is the root of the problem. an increased trend to outsourcing shifts large portions of the information technology systems outside the company control and responsibility. in some ways. Everyone wants to know more about it. For instance. Virtually anyone with an interest in information technology has one.related capabilities are provided ‗as a service‘ across the Internet to multiple external customers‖ 6 . the propensity to adopt may be reversed from other technologies. All the vendors are rebranding their products as being aligned with the cloud. technologically savvy organizations prefer new developments while those who like to remain distant from technology also like to keep the status quo as long as possible. In fact. Typically. which they never considered to be contributing to the bottom line. In the case of cloud computing. you will have trouble building a solid architecture if you are unsure what you are trying to build.

One popular analogy is an old Indian story of six blind men who encountered an elephant and attempted to determine what it was using their hands. The most important consideration for any IT manager is not whether a potential solution satisfies the definition of cloud computing but rather whether it adds value to the business. they all do a good job at giving an idea of what is involved. highly scalable. they are still not contradictory. Users need not have knowledge of. This observation may be somewhat discouraging but before you concede defeat you might consider that. I won‘t add to the confusion with another attempt at perfecting the definitions that are already available. or in some cases even very similar. although the definitions are not identical. And it‘s not obvious to me that there is any particular value in having an authoritative definition. the trunk felt like a tree branch. Each description held an element of truth and yet fell short of a complete description. and the ear like a hand fan. most definitions of cloud computing include elements of the complete description and yet they typically do not address every single aspect that anyone has associated with cloud computing. And 7 . expertise in. They simply emphasize different aspects of a complex and multi-faceted notion. A cloud-based solution that does not increase revenue or decrease costs is of little interest. Similarly. another felt the tail which he likened to a rope.‖ This is a comprehensive definition but it hides the primary discovery of the paper that no single theme seemed to permeate all of the definitions. One felt a leg and compared it to a pillar. and managed infrastructure capable of hosting end-customer applications and billed by consumption‖ Wikipedia: ―A style of computing in which dynamically scalable and often virtualized resources are provided as a service over the Internet. For the most part.Forrester: ―A pool of abstracted. or control over the technology infrastructure "in the cloud" that supports them. the belly was like a wall while the tusk felt comparable to a pipe.

combined with on-demand service provisioning. Since utility billing is also common this elasticity can equate to direct cost savings. This doesn‘t mean that every cloud attribute is essential to cloud computing or even that there is necessarily any which qualifies a given approach as fitting the cloud paradigm. which is made available to the end-user. In other words. 8 . But typically the more of these attributes apply. The model goes much further in providing an elastic provisioning mechanism so that resources can be scaled both up and down very rapidly as they are required. Monetization can take the form of placed advertising or can rely on simple credit card charges in addition to elaborate contracts and central billing.a completely non-cloud-oriented solution that does unambiguously improve the bottom line should be implemented regardless of the name by which you call it. the service must cross both physical and security boundaries. ATTRIBUTES The key to understanding common interpretations of the term ―Cloud Computing‖ is to examine the assortment of attributes of typical cloud solutions. of resources. Fees can be levied on a subscription basis or can be tied to actual consumption. facilitate a number of options for charging customers. they are neither necessary nor sufficient prerequisites to the notion of cloud computing. Flexible Billing: Fine-grained metering or resource usage. This usually has two implications: The service is delivered over the public Internet and the processing occurs outside the company firewall. On their own. the more likely others will accept it as a cloud solution. or reservation. Elasticity: One main benefit of cloud computing is the inherent scalability of the service provider. Some key components include: Off-Premise: The service is hosted and delivered from a location that belongs to a service provider.

critical applications to the extent that others (which are governed by more precise commitments) might. location independence and high levels of resilience allow for an always-connected user experience. They leverage various virtualization mechanisms and achieve cost optimization through multitenancy. Service-Level Management: Cloud services typically offer a Service Level definition that sets the expectation to the customer as to how robust that service will be. Service Delivery: Cloud functionality is usually available as a service of some form. While there is great variance in the nature of these services. 9 . user self-service to expedite business processes and programmatically accessible resources that facilitate integration into enterprise management frameworks. typically the services offer programmatic interfaces in addition to the user interfaces. There is no requirement for fixed purchases. Pooling resources across customers is an important factor in achieving scalability and cost savings.Virtualization: Cloud services are usually offered through an abstracted infrastructure. And the economy of scale of the service providers allow them to optimize their cost structure with commodity hardware and finetuned operational procedures that are not easily matched by most companies. Multi-tenancy: The cloud is used by many organizations (tenants) and includes mechanisms to protect and isolate each tenant from all others. Universal access: Resource democratization means that pooled resources are available to anyone authorized to utilize them. At the same time. Simplified management: Administration is simplified through automatic provisioning to meet scalability requirements. Affordable resources: The cost of these resources is dramatically reduced for two reasons. They can still be considered cloud services but typically will not be ―trusted‖ for mission. Some services may come with only minimal (or non-existent) commitments.

Service-Oriented Architecture A service-oriented architecture (SOA) decomposes the information technology landscape of an enterprise into unassociated and loosely coupled functional primitives called services.RELATED TERMS In addition to the set of characteristics which may be associated with cloud computing it is worth mentioning some other key technologies that are strongly interrelated with Cloud Computing. The best way to think of the relation between them is that SOA is an architecture which is. However. the two are independent notions. However. they have also been applied 10 . They are usually confined to scientific problems which require a huge number of computer processing cycles or access to large amounts of data. by nature. One of the greatest advantages of this approach is that it maximizes reusability of functionality and thereby reduces the effort needed to build new applications or modify existing programs. These grids are often based on loosely coupled and heterogeneous systems which leverage geographically dispersed volunteer resources. The business logic is then tasked with orchestrating the service objects by arranging them sequentially. selectively or iteratively so that they help to fulfill a business objective. Cloud computing may also drive increased attention to SOA. There is a high degree of commonality between cloud computing and SOA. In contrast to monolithic applications of the past. Cloud Computing may be one means of implementing an SOA design. technology independent. Grid Computing Grid Computing refers to the use of many interconnected computers to solve a problem through highly parallel computation. An enterprise that uses a service-oriented architecture is better positioned to leverage cloud computing. these services implement single actions and may be used by many different business applications.

Darcy DiNucci first used the expression in 1999(DiNucci. The ultimate objective has been on the radar screen of many IT companies (such as Sun. and the absence of legacy dependencies. But it is important to also be aware of their distinctions.0 The term Web 2. This term has evolved to refer to the web as not only a static information source for browser access but a platform for web-based communities which facilitate user participation and collaboration. Web 2. Tim O‘Reilly popularized a recast notion in his Web 2.0 is one of the fastest growing areas for new applications and that it typically involves little infrastructure other than the computers and networks needed to make the services available to users on the public Internet. distribute their workload and blur the line between system usage and system ownership. Cloud services are typically opaque and cover a wide range of almost every class of informational problem using a model where the functionality is decoupled from the user. There is no intrinsic connection between cloud computing and Web 2.0 conference in 2004(O'Reilly & Batelle. A grid may be transparent to its users and addresses a narrow problem domain. 2004). These requirements.0. Both involve large interconnected systems of computers. it is worth observing that Web 2.0 is also often closely associated with cloud computing. economic forecasting.successfully to drug discovery. seismic analysis and even financial modeling for quantitative trading including risk management and derivative pricing. There may be some conceptual similarity between grid and cloud computing. 1999) to refer to radical changes in web design and aesthetics. HISTORY Cloud computing represents an evolution and confluence of several trends. Cloud computing is a means of delivering services and Web 2. Nonetheless. HP and IBM) for 11 .0 is a class of services which may be delivered in many different ways. make it optimally suited to cloud platforms.

with other retailers in 12 . Amazon started as an on-line bookstore in 1995. apparel and even automotive parts and accessories. The approach met their needs however it was not inherently optimized. A major change in business model involved the creation of merchant partnerships that leveraged Amazon‘s portal and large customer base. As Amazon grew. grocery. they had several other advantages that they were able to leverage effectively. it had to find ways to minimize its IT costs. But the first commercially viable offerings actually came from other sectors of the industry. depending on how you look at it. As most readers will recall. However. Amazon chose to minimize hardware expenditures by buying only commodity hardware parts and assembling them into a highly standardized framework that was able to guarantee the resilience they needed through extensive replication. Without the bricks-and-mortar retail outlets. In the course of building their infrastructure. Amazon brokered the transaction for a fee thereby developing a new ecosystem of partners – and even competitors. Its business model (Amazon received 615 million visitors in 2008 according to Wikipedia) implied a very large online presence which was crucial to its success. eventually expanding into computer hardware and software. This may seem somewhat odd since Amazon was not initially in the business of providing IT services. its data center investments and operations became a significant portion of its cost structure. their system designers had scrutinized the security required to ensure that the financial transactions and data of their customers and retail partners could not be compromised. Amazon was arguably the first company to offer an extensive and thorough set of cloudbased services. DVDs. Based on its success in the book market it diversified its product portfolio to include CDs. Amazon and partners shared a common burden or boon. jewelry. and other forms of digital media.several years.

Also keep in mind that there are several other important cloud service providers such as Salesforce. In order to satisfy hundreds of millions of search requests every day they must process about one petabyte of user-generated data every hour(Vogelstein. their primary business model is fundamentally different in that they do not have a huge retail business which they can leverage to easily monetize their services. they effectively sold some of their idle capacity to other organizations who had computational requirements from January to November.that a very high proportion of their sales are processed in the weeks leading up to Christmas. I will not go into detail on the different implications of these two approaches at this point but it will be useful to keep in mind as we discuss the various cloud platforms in the next chapters. When they launched Amazon Web Services in 2002. In order to be able to guarantee computing capacity in December they needed to overprovision for the remainder of the year. Through extensive data mining they are able to identify and classify user interests. Their story bears some resemblance to Amazon‘s. Instead Google‘s source of revenue is through advertising and their core competence is in analytics. The inefficiency contributes to an unacceptable amount of unnecessary costs. The proposition was attractive to their customers who were able to take advantage of a secure and reliable infrastructure at reasonable prices without making any financial or strategic commitment.com or Microsoft. Each has their own history and business model. However. This meant that a major share of their data center was idle eleven out of twelve months.2009). given the success of Google Apps and Google App Engine. The two above are merely two 13 . Google is another player in Cloud Computing history that is worth examining. They also host a huge worldwide infrastructure with many thousands of servers worldwide. And through their portals they can place advertising banners effectively. Amazon decided to turn their weakness into an opportunity.

Amazon is constantly enhancing the EC2 offering in an effort to increase differentiation and add value. Google App Engine is generally considered to be a Platform as a Service. this is a simple task but it is difficult to aggregate the functionality in an optimal manner unless you can get a clear picture of all the services that are available. or investing in single-purpose appliances. And Salesforce. Over time. CLOUD ARCHITECTURE STACK One characteristic aspect of cloud computing is a strong focus toward service orientation. As is often the case with classification systems. In principle. services may also drift between service types. For example. That is not to say that it could not be completely different players who shape the future. which can be assembled as required. the lines are not nearly as clear in reality as they may appear on a diagram. As the product matures. This is a lot easier if you can provide some structure and a model that illustrates the interrelationships between services. Rather than offering only packaged solutions that are installed monolithically on desktops and servers. The most common classification uses the so-called SPI (Software as a Service. Amazon Elastic Compute Cloud (EC2) is a classical example of Infrastructure as a Service. There are many services that do not fit neatly into one category or the other.com represents one of the best known examples of Software as a Service. some may begin to question if it wouldn‘t be more accurate to consider it a platform service. Platform as a Service and Infrastructure as a Service) model.notable examples which I feel provide some insight into the history of cloud computing. you need to decompose all the functionality that users require into primitives. 14 .

At the other extreme. they can only accommodate minimal customization and extensions. infrastructure services can host almost any application but are not able to leverage the benefits of economy of scope as easily. it is easiest to begin with a conceptual distinction. Platform services represent a middle ground. The classification illustrates how very different these services can be and yet. However. at least conceptually. each layer depends on the foundation below it ( Figure 1-5) Platforms are built on infrastructure and software services usually leverage some platform 15 .Figure 1-4 Software. There are two primary dimensions which constrain the offerings: The services differ according to their flexibility and degree of optimization (Figure 1-4). Software services are typically highly standardized and tuned for efficiency. Platform and Infrastructure services Nonetheless. They provide flexible frameworks with only a few constraints and are able to accommodate some degree of optimization.

Figure 1-5 SPI Model Figure 1-6 SPI Origin Figure 1-7 Extended Model 16 .

it‘s useful to take a look inside (Figure 1-7). In the basic scenario the datacenter owner rents out floor space and provides power and cooling as well as a network connection. Platforms represent frameworks and common functions that the applications can leverage so that they don‘t need to re-invent the wheel. and libraries with interfaces to frequently needed functions. backup and recovery. One of the most popular areas is customer relationship management (CRM). There are also platform services that focus on specific components such as databases. In fact. human resources. identity management repositories or business intelligence systems and make this functionality available to application developers. development environments. legal. they lend themselves well to further segmentation. There are a wide range of domains where you can find SaaS offerings. There are providers of simple co-location (facilities) services. I don‘t mean to imply that they are any more complex or diverse than platform or software services.We will examine each of the layers in more detail in the next three chapters. as well as forms of collaboration (such as conferencing or unified communications). I have divided infrastructure services into three sublevels. and many other domains appearing regularly on the market. But to give you an idea of what each represents. 17 . But the list is endless with services for billing. they are probably more homogenous and potentially even simpler than the higher tiers. Software Services represent the actual applications that end users leverage to accomplish their business objectives (or personal objectives in a consumer context). I suggest that most infrastructure services fall into three categories that build on each other. However. financials. The rack hardware may also be part of the service but the owner is not involved in filling the space with the computers or appliances that the customers need. Desktop productivity (including electronic mail) is also very common. The offerings often include programming language interpreters and compilers.

Very often this implies that they will install a virtualization layer that facilitates automated provisioning. Figure 1-8 Implementation. MANAGERMENT LAYER In addition to the software and applications that run in the SPI model and support a cloud application in its core functions. It is 18 . The simplest options involve dedicated servers. There are also managed hosting providers who will take over the administration. they will leverage multitenancy in order to maximize resource utilization . there are also a number of challenges that both the enterprise and service provider need to address in order to successfully keep the solution going (Figure 1-8). internal networking and storage equipment that is operated by the customer. There are a large and ever increasing number of cloud-based services and solutions on the market. There are hosting services that will provide and install blade systems for computation and storage. resource management and orchestration while also enforcing consistency of configuration. monitoring and support of the systems.but this is not strictly required. In some cases. Operation and Control It is necessary to select and integrate all the components into a functioning solution.The next conceptual level is to add hardware to the empty rackspace.

Operate Once the solution has been brought online it is necessary to keep it running. track incidents and implement changes in the service. AND HYBRIDS CLOUDS Cloud computing broadly breaks down into 3 different models . PRIVATE. PUBLIC.choosing the right one starts with understanding business objectives and how best they align with the values each model delivers. plan for capacity changes. decommission old users. Some of the challenge may be reduced by working with solution providers and outsourcing organizations who take over the operative responsibilities. When weighing up which Cloud platform to use. It is still necessary to ensure that service expectations are well defined and that they are validated on a continuous basis. And once that is done. Also consider any regulatory or data protection requirements. Since the service is unlikely to be completely static you need to also have processes in place to provision new users. and how integrated applications need to be with the other enterprise functions. troubleshoot it and support it. The integration effort involves a careful selection of interfaces and configuration settings and may require additional connectors or custom software. This means that you need to monitor it. this doesn‘t completely obviate the need for overseeing the task. However. it would be naïve to expect them all to work together seamlessly. Control The operation of a complex set of services can be a difficult challenge. All 19 . as well as application workloads.no simple task to categorize and compare them. businesses need to consider first and foremost how critical the applications they wish to move into the cloud are to their business.

the business can simply sign-up for and start using Cloud Computing. Private or Hybrid Cloud solution is best suited. as they are effectively ‗stand- 20 . whether managed internally or hosted by a third-party externally. high availability or fault tolerant solutions that are not possible in a Public Cloud. Under this scenario. Software as a service (SaaS) and Desktop as a service (DaaS) platforms – all of which we make available to the channel through our proprietary Cloud services. Private Clouds offer scope for advanced security. The economies of scale afforded by Public Cloud computing are what make this technology highly attractive. Public clouds A Public Cloud is one that‘s based on the standard cloud computing model where services. They require a significant level of engagement from both management and IT departments to virtualize the business environment. the most common being Infrastructure as a service (IaaS). applications and storage are made available to users over the Internet ‗as a service‘ – typically on a Pay Per Use model. Private cloud Private Clouds consist of cloud infrastructure that is deployed solely for a single organization. there is less scope for customization for security and performance. There are many types of Public Cloud. However. online storage and other services immediately. or are looking to outsource part or all of their organizational IT requirements. Platform as a service (PaaS). have less regulatory hurdles to overcome. Suited to – Companies that need to bring a service to market quickly. While Public Clouds are appealing to many businesses as they reduce complexity and lead times.of these factors help determine whether a Public. because the underlying architecture is fixed. and also mean evaluating how existing resources should be reallocated in the cloud.

a hybrid cloud comprises both private (internal) and public (external) cloud services. businesses install their own server and storage hardware but have the flexibility to shift workloads among servers as usage spikes or they deploy new applications. Suited to – E-commerce. and this type of sensitive data is more secure if it stays ―on-premise‖ in the Private Cloud. legal regulations strictly govern how personal and payment information can be handled. Typically a business might run an application primarily on a private cloud.alone‘ solutions in their own right. building a Private Cloud still involves significant investment. Suited to – Businesses that must comply with strict regulations or that have highly critical applications will choose internal Private Clouds. Customized rules and policies govern areas such as security and the underlying infrastructure. Because e-commerce sites must respond to fluctuations in traffic both on a daily and seasonal cycle. the actual work of processing the orders can benefit from the elastic characteristics of public cloud resources. but rely on a public cloud to accommodate spikes in usage. and does not therefore deliver the shorter-term economies that Public Cloud can. with tasks allocated to internal or external clouds as necessary. This hybrid 21 . Table 1-1 Private and Public Cloud Hybrid clouds As the name suggests. On the other hand. With a private cloud.

which segregate the client from the business logic and both from the information and database layer (Figure 1-9). 22 . TOPOLOGY Over the past half century we‘ve seen the typical computer topology shift from the mainframe in the 1960s to the client-server computing in the 1980s. In particular it is very important to consider the client device as part of the complete cloud computing topology. Figure 1-9 Connectivity Evolution We are now seeing an increase in mesh connectivity.solution represents the best of both worlds as it places the order processing and transactional front-end of the shop where it can take advantage of resource scalability. while it keeps the payment and account management services strictly private. peer-to-peer networks leverage the fact that every system on the network can communicate with the others. The 1990s popularized the notion of N-tier architectures. For example. Data processing and storage may be shared between systems in a dynamic manner as required. Desktop virtualization can have a fundamental impact on cloud computing and can also leverage cloud services to provide content on the terminal. Cloud computing can facilitate any of these models but is most closely associated with a mesh topology.

the user may then receive the content directly from the solution provider but. Some functionality is simply easier to process locally.However. The entity that creates the content or provides the ultimate functionality may be hidden from the user. it is often the case that the provider partners with a content delivery network. Instead the user interacts primarily with a content aggregator who bundles the services and contents into a form that add value to the user. There are many possible options for delivering content on the Internet. In some cases. Desktop applications. Therefore it is not realistic to think that cloud equates to thin client computing. We may have reached limits in transistor density but processing power is still advancing with multi-core processors. may be more suitable for the cloud. Moore‘s law continues to apply. particularly those that are collaborative in nature. Figure 1-10 Content Delivery Model There are at least three different players in many solutions. These network providers have extensive global presence and very good local connectivity. particularly for high-volume and latency/bandwidth sensitive content. such as Akamai or Limelight. It is inherent in a serviceoriented architecture that the end user not be explicitly cognizant of the individual component services. Content Delivery Model: One way to look at topology is to trace the content flow. may seamlessly extend into the cloud. with high computational requirements. The key challenge ahead will be the effective synchronization and blending of these two operating modes. These do not necessarily change through cloud computing. We may also see more potential for hybrid applications. They can 23 . such as MATLAB and Mathematica. while other functions. but it is important to be aware of all actors and their respective roles since they are all very much a part of cloud offerings too.

has made the sector extremely attractive in an economic downturn. There are two key parts of the cloud computing ecosystem that you should keep in mind as you look at different offerings. There are literally hundreds of vendors who consider some of their products and services to relate to cloud computing  It is very dynamic.  It is extremely large.replicate static content and therefore make it available to end-users more quickly. the roles of the value chain are more complex and diverse than just described. ECOSYSTEM In reality. The hype surrounding cloud computing. Value Chain: Although there is some correlation. This means there are many players entering the market. An ecosystem ties together a fragmented set of cloud computing vendors. But it also means that many are dynamically reshaping their offerings on a frequent basis often extending into other cloud areas. This is intuitive and merely reflects a means of profit sharing toward those who facilitate the end-to-end service. thereby improving the user experience and off-loading the hosting requirements from the aggregator. combined with the lack of entry barriers for many functions. Figure 1-10 Payment Ecosystem The simple part of the payment model is the flow from the aggregator to the delivery network and content creator. the path of content delivery is quite distinct from the payment and funding model (Figure 1-10). Even the delivery mechanisms 24 . Some are exiting.

2008). Open Source and Cloud Computing Richard Stallman. be delivered as a service. over the Internet. business intelligence and a variety of other areas. it is very difficult to paint an accurate picture of the ecosystem which will have any degree of durability or completeness to it. His concerns around loss of 25 . information management. there are services for integrating. A less obvious form of impact is that each of the functions needed to implement cloud computing can. It is unclear where this path will ultimately lead and whether all computational assets will eventually be owned by a few service providers. Slogans such as ―Anything as a Service‖ or ―Everything as a Service‖ are becoming more popular to indicate that we not only have software. and priced to reflect the actual use and value to the customer. There are also emerging services for printing.themselves are changing as the technologies evolve and new functionality becomes available. But the trend is certainly in the direction of all functionality that is available also being accessible on-demand. a well-known proponent of open source software attracted significant publicity for his skepticism of cloud computing (Ricciuti. platforms and infrastructure as services. On top of these. storage and security. The market is changing and I can only provide a glimpse and high-level overview of what it looks like at this point in time. A direct outcome of this diversity is that we can expect the effects of the technology to cross many boundaries of influence. managing and governing Internet solutions. but also components of these such as databases. leveraged by end users only if and when they need them. Total Cloud There are many parts to cloud computing and each of these components can be technically delivered in many different ways using a variety of different business models. itself. which can be offered on-demand and priced on a utility basis. As a result.

Distributed cloud frameworks. Many SaaS solutions leverage open-source software for obvious cost and licensing reasons. Cloud providers often use Xen and KVM for virtualization to minimize their marginal costs as they scale up. SaaS is an ideal monetization model for open source since it facilitates a controlled revenue stream without requiring any proprietary components. Time will tell whether enterprises and consumers take advantage of this shift to reduce their desktop license fees by adopting Linux. such as Hadoop. Even though some synchronization will always be useful. cloud-computing providers will depend on open. For example. (Gartner. It is an evolution of virtual private server offerings and merely provides a mechanism to take advantage of 26 .control and proprietary lock-in may be legitimate. In summary.source software to deliver products and services. 90% of market-leading. Google Android or other open-source clients. are usually open source to maximize interoperability and adoption.. it is also interesting to observe that cloud computing leverages open source in many ways. Gartner has projected in one of their Strategic Planning Assumptions that through 2013. there is the potential that cloud computing may act as a catalyst for open source. SERVICE MODELS OF CLOUD COMPUTING INFRASTRUCTURE AS A SERVICE ( IaaS) Infrastructure as a Service (IaaS) is the simplest of cloud offerings. Web-based APIs also make the client device less relevant. In some ways. Self-supported Linux is by far the most popular operating system for infrastructure services due to the absence of license costs. Nonetheless. the value proposition of thin clients increases as the processing power and storage shifts to the back end. SugarCRM is one of the most common platforms for cloud-based CRM. 2008). Inc.

Subterranean placement of the data 27 . Infrastructure Stack In order to understand infrastructure services it is useful to first take a look behind the scenes at how an Infrastructure Service provider operates and what it requires in order to build its services. an outage in either one can disrupt the entire operation of the facility and cause serious damage to the equipment. After all. it is necessary to have uninterruptable power supplies or backup diesel generators that can keep the vital functions of the data center going over an extended period of time. Points of intersection between the electrical grids of regional electricity providers are particularly attractive since they facilitate a degree of redundancy should one utility company suffer a wide-spread power outage. Power and cooling are critical to the functional continuity of the data center. More importantly.hardware and other physical resources without any capital investment or physical administrative requirements. Another environmental requirement is an efficient cooling system. they can represent over a third of the entire costs so designing them efficiently is indispensible. The benefit of services at this level is that there are very few limitations on the consumer. Often drawing multiple megawatts. Over half of the power costs of a data center are often dedicated to cooling. There may be challenges including (or interfacing with) dedicated hardware but almost any software application can run in an IaaS context. It is very important for the data center to have access to multiple power sources. As costs have sky-rocketed there have been numerous advances in cooling technologies and techniques. In any case. Most recent cooling designs leverage outside air during the colder months of the year. the tasks and the challenges of the provider are directly related to the benefit of the customer who is able to outsource the responsibilities.

center can lead to better insulation in some parts of the world. The cabinets are designed according to the Electronic Industries Alliance EIA-310D specifications which designates dimensions.45 mm) high. 44. log data) and plenum spaces with air circulation underneath the floor. The interior of the data center is often designed to optimize air flow. rack openings and other physical requirements. Most racks are available in 42U form (42 x 1. These are typically organized in rows of servers mounted in 19-inch rack cabinets. 28 . Whoever owns and operates the data center must also come up with an internal wiring plan that distributes power and routes network access across the entire floor wherever computer hardware or other electrical infrastructure is likely to be placed. This aggregation can lead to increased operational efficiency in that less manual labor is involved and the possibility of human error is reduced. an advantage of the larger enclosures is the reduction in number of external wires that are necessary since much of the switching fabric is internalized to the system. Each cabinet accommodates modules which are 19 inches (480mm) wide and multiples of 1U (1. These modules then need to be wired for power and connected to the network. The next step of an infrastructure provider is to fill the rented (or owned) data center space with hardware. targeted vents (using sensors. storage units and network appliances that can be accommodated in the cabinet. The challenge is to maximize the number of servers.5 inches high) so the space is limited. HP BladeSystem). for example through alternating orientation of rows of racks.75 inches. which can host larger numbers of microblades than the U slots which they occupy.75 = 73. Alternatively. server vendors such as HP and IBM offer enclosures (IBM BladeCentre. Again. But the density can be augmented by increasing the proportion of 1U blades versus 2U and 3U rack-mountable components. hole spacings.

system. 29 . Memory virtualization can abstract volatile memory space and map it to a set ofpooled memory resources among networked systems. or virtual. sometimes called cells. The abstraction of the physical infrastructure can facilitate higher utilization through pooling of units and thin provisioning. or streaming of the operating system and/or applications. Storage virtualization involves redirecting I/O requests from logical. Server virtualization abstracts the underlying physical resources and presents these as a set of virtual machines. They may be pre-loaded. Desktop virtualization is a term that embodies yet again a number of related concepts ranging from thin-client computing to application encapsulation. operating systems) and the means by which they are delivered. that act as their own self-contained infrastructure. These bundles of servers on dedicated networks contain their own management and security components to operate a fully functional. loaded at boot time. But certainly the most prominent in the cloud context is server virtualization.There has also been extensive research into providing clusters of functionality. Server virtualization . It also makes it much easier to migrate data without disrupting a service. redundancy and better utilization. Again. units to physical storage devices. streamed as needed or simply hosted remotely and presented on the desktop. this offers flexibility. each of which appears to its applications and users as though it were a physical system. The applications can continue to make the same logical requests even if the data is transferred to another device. These delivery models vary according to the degree of isolation they provide (applications. containers. and often complex.The virtualization techniques stated above all offer benefits to cloudbased services.

2010). Nonetheless. This encapsulation makes it very easy to load-balance and redeploy applications as usage requires. There have been many kernel and hypervisor bugs and patches. Probably the most important driver is the fact that virtual machines can be launched from a virtual disk independent of the hardware on which they were configured. Fine-grained instrumentation provides the foundation by accounting for usage and delivering accurate information that can be used for internal cross-charging or fed into a billing and payment system for external collection. A study by Greenpeace picked up alarming trends on the growing carbon footprint of cloud data centers and cloud-based services (Greenpeace. Cloud computing primarily shifts the location of processing and storage. And it makes it very easy to provision new instances instantly when they are needed. the economy of scale of the cloud service providers can also help to address environmental objectives more effectively. the hypervisor typically presents a smaller attack surface than a traditional operating system and is therefore usually considered superior to application isolation. It also enforces a level of standardization in configuration between similar instances of the same application. However. It is in the provider's interests to minimize energy 30 . which doesn't directly translate to a global ecological benefit. which might be compromised through kernel exploits in the host operating system.Server virtualization provides a high degree of isolation between guest operating systems. It is simply a matter of copying the virtual machine to a new machine which is running the same hypervisor. This doesn‘t mean that it is immune to attack or vulnerabilities.

expenditures and maximize reuse of equipment. Physical allocation means that specific hardware is allocated to the customer as in the examples above. Its scale also increases the attractiveness of investments in sophisticated power and cooling technology. Cloud computing can be seen as the evolution of managed hosting providers such as Navisite. They offer co-location capabilities as well as dedicated pools of rack-mounted servers. Beyond these distinctions. Shared virtual servers are exposed to the customers as pools of virtual machines. elaborate systems of temperature sensors and air flow ducts as well as analytics based on temperature and energy logs. Dedicated virtualized servers offer dedicated hardware but provide a hypervisor on the physical machine so that the customer can run multiple operating systems and maximize server utilization. Terremark or Savvis. which it can achieve through higher consolidation of infrastructure. Their managed hosting capabilities often include virtualized resources on dedicated infrastructure with console. the key differentiating options are the operating systems that are supported (usually confined to Windows and a set of Linux distributions) and the packages that are available off-the-shelf (e. 31 .g.based provisioning. application frameworks). Shared Virtual. The server outsourcing model can be divided into three allocation options: Physical. including dynamic smart cooling. It is not discernible on which physical equipment a particular instance is running or what other applications may be co-resident on the same machine. Dedicated Virtual. Infrastructure Services Services represent the allocation of computational resources along with minimal storage and input/output channels.

One challenge in using many virtual servers is that they typically do not maintain any local storage and may even lose their state between invocations. The Rackspace Cloud (formerly called Mosso) is another very well known 32 . or easier.g. They employ VMware and 3Tera Applogic for the virtualized solutions along with a proprietary technology called Virtuozzo Containers that differ from conventional hypervisors in that they offer isolation of virtual machines while sharing the underlying operating system. operating systems and packages. Each virtual machine is based on an Amazon Machine Image (AMI). any important configuration information or other data can be stored externally.  LayeredTech provides co-location and dedicated as well as virtual server offerings. local disk size). Some of the best known include:  Amazon falls into the category of shared virtual machines and is perhaps best known. Some server offerings do exactly that. JBoss. But for some purposes it may be better. There are advantages to this approach. They vary in resources (RAM. operating systems (Several Windows versions and many Linux distributions) and the application frameworks that are installed on them (e. The customer can use pre-packaged AMIs from Amazon and 3rd parties. Their differentiation lies in the degree of visibility they provide on the location of the server with transparency of the data center as well as rack location and position. Oracle). MySQL.  AppNexus offers dedicated virtualized servers based on Dell computers with Ipsilon storage.There are a variety of providers that offer multiple combinations of allocation options. Compute units. or they can build their own. to maintain persistent local storage. After all.

MySQL. KVM and other virtualization solutions. <=. >. Fedora. UNION. >=. It is common for these services to be able to also accommodate unstructured data (blobs). Instead it defines ―domains‖ with items that consist of up to 256 attributes and values. there are third party and native components for Xen. trustworthy payment system. So most common queries are possible as long as they are confined to a single domain.IaaS provider with many managed hosting options in addition to virtual server offerings covering a number of Linux distributions (such as Ubuntu. OR. INTERSECTION. Effective monetization of utility-based services relies on two capabilities: fine-grained metering and a simple. Centos and RedHat Enterprise Linux). Joyent also offers a private version of their framework called CloudControl for enterprise data centers. They run OpenSolaris with Apache.installed and the ability to add other packages. In the end. The values can contain anywhere from one byte to one kilobyte. The Amazon Simple DB (SDB). Note that even though most of the information is tabular it is typically not SQL-conformant and may not support joins. It has a large pool of dedicated IP addresses and offers persistent storage on all instances. There are also cloud database options for more structured data. !=. It also supports simple operators such as: =. NOT. Nginx. <. Depending on the hypervisor in use. triggers or stored procedures. AND. Arguably. PHP. Hyper-V. most infrastructure-layer components are managed through the facilities of the virtualization vendor. Joyent uses the term ‗accelerators‘ to refer to its persistent virtual machines. STARTS-WITH. foreign keys. A feature called ―automatic CPU bursting‖ provides reactive elasticity. Ruby on Rails and Java pre. VMware. for example does not use a schema. the leaders in these domains 33 .

One of the most frequent uses of IaaS is as a SaaS enabler. 34 .have been the telecommunications companies and. To understand this multi-level business model it is important to keep in mind that while there is a significant distinction between infrastructure services and software applications this doesn‘t mean they are unrelated. However. but may also make available to their customers and partners for a fee. infrastructure services tend to run on platforms that were designed for desktops and traditional client-server environments. At the same time. In addition to providing the raw computing and storage power some IaaS providers also help to facilitate SaaS with payment and billing services that the customer can leverage to monetize their SaaS services PLATFORM AS A SERVICE ( PaaS) Platform as a Service has emerged from a confluence of two trends: The suboptimal nature of IaaS for cloud computing and the evolution of web applications. eCommerce has also boomed in recent years based primarily on credit-card billing. It is therefore no surprise to see cloud offerings appearing from international communications giants such as BT. However. such as PayPal. Companies. Infrastructure service offers many benefits to customers who wish to extend or shift their applications into a cloud-based environment. make themselves available as trusted intermediaries for smaller merchants that do not yet have the scale and visibility to establish public trust. They may now be virtualized but they have not been optimized for the cloud. the mobile operators. DT. which they can use for their own business. many cloud service providers have developed their own instrumentation and billing systems. in particular. Orange and SingTel. NTT.

C#:ASP. While a simple web service only needs to serve HTML over HTTP it is very cumbersome to build complex and dynamic web sites that do not utilize some level of server-side business logic built on the information contained in a data repository.NET is based on Microsoft ‘s Active Server Pages (ASP) technology. In addition to in-line mark-up. which consist of HTML as well as dynamic code that is pre. Eiffel. Ada 35 .NET code into a separate file that is only referenced in the .NET languages such as Visual Basic and C#.To better address the specific needs and advantages of cloud delivery models. it is possible to separate the . In order to facilitate the necessary functionality.NET pages.NET is a component of Microsoft Internet Information Server (IIS) and is closely connected with Microsoft Visual Studio. Ruby: Ruby on Rails is an open-source framework that supports Ruby. and an easier development environment. page templating.processed when the page is rendered. database access. Smalltalk.NET. While ASP. it is also possible to extend it with other frameworks such as DotNetNuke.NET page. which is compatible with all Microsoft . authorization. Castle Monorail or Spring. some vendors have crafted new platforms that enable faster time-to-market. revised to leverage the Common Language Runtime (CLR). It leverages . a common user experience. server-side caching and session management. Web Application Frameworks Let‘s begin with a historic look at web hosting. These are often classified according the programming languages that they support: Visual Basic. a dynamic and reflective object-oriented programming language that is based on Perl. a number of application frameworks have developed that alleviate the overhead of authentication.

Python: Python is characterized by its support of multiple programming paradigms including object-oriented. It is used in a variety of popular frameworks including Catalyst. not all Java environments are identical. typically acting as a filter that processes input data and renders HTML. 36 . However.and Lisp. structured. Web Hosting Services The boundary between conventional web hosting services such as 100Webspace. it suffered in equal measure when Twitter switched to Scala due to scalability issues. The best-known include Apache Struts and Spring Framework. which have delivered thousands of modules that enable developers to build extensive content management systems without any custom coding. AwardSpace. Perl: Perl is a general-purpose programming language that was originally developed for manipulating text but is now used for a variety of applications including system administration and web development. Jifty and WebGUI. functional and aspect-oriented programming. However. PHP: PHP was originally developed for building dynamic web content. X10Hosting or HelloHost and a Platform as a Service provider is blurred. Application frameworks such as Drupal and Joomla have spawned popular ecosystems. Ruby on Rails received considerable attention as the original basis for Twitter. Java: Java requires little introduction as it is a de facto standard for open-source software. There are several Java application frameworks.

In addition to simple web file services they support a number of the application frameworks mentioned above. Apache. On the other hand.The former offer hosting services using business models that range from placed advertisements to one-time setup fees or monthly subscriptions. Most experts would not consider these sites to be PaaS providers simply because the provisioning support and value added services of the platforms are minimal. and ideally presents these to the end user in a fashion that is not radically different from how users expect to use their applications. which should be invisible to corporate endusers. The services run on the provider‘s infrastructure and are accessed through a public network connection. SOFTWARE AS A SERVICE ( SaaS) Software as a Service (SaaS) differs from Infrastructure and Platform as a Service in that it provides a service which is directly consumable by the end-user. Applications may be made available through Internet as browser applications or they may be downloaded and synchronized with user devices. licensing and billing model. SaaS is a model whereby the customer licenses applications and provisions them to users on demand.NET or a complete LAMP (Linux. The Rackspace Cloud (primarily known for its infrastructure services) offers a service called Cloud Sites. IaaS and PaaS offer infrastructure and platforms where system managers and developers can install their applications but they offer little intrinsic value to a non-technical user. 37 . There may be some user interface changes that ripple through to the users but the main difference is the deployment. Consistent with the basic notion of cloud computing. MySQL. SaaS provides the full stack of cloud services. Perl/Python/PHP) stack and would typically qualify as a PaaS offering. These fully managed platforms can host Windows . Mosso.

which lowers fixed costs and capital investment. In fact. The most common pricing model is based on the number of users but there may be additional fees based on bandwidth. It simplifies licensing. the customer doesn‘t need to acquire (or directly pay for) a software license at all. storage and usage. SaaS is a standardized and optimized solution that generally will result in the lowest costs to the customer. Rather than theorizing about their composition it is more effective to provide an overview of the types of software that they represent. However. As long as the functionality offered by the vendor exactly matches the requirements of the customer. Financial There are a variety of financial applications that are available on-demand. 38 . This is a task of the provider. It outsources the tedious task of application maintenance and upgrades and ties customer costs to usage. SaaS offers several compelling benefits. It is challenging to provide any abstraction to SaaS services or to classify them according to the technology and approach that they use. the flexibility to customize. ranging from accounting to procurement and inventory management. There is also no need to calculate maximum capacity. the range of business sectors they span and the functionality that they can deliver and scale on demand. They are incredibly diverse and can leverage any web-based tools to implement the application requirements. or indeed to build additional applications. is much greater with PaaS or IaaS. Typically. they are highly standardized but may vary in their configurability as well as their efficiency and scalability. SaaS has both advantages and disadvantages compared to PaaS and IaaS.Some of the characteristics of SaaS services are that they are centrally managed and updated.

 Financial accounting: for investors and creditors as well as to ensure SarbanesOxley compliance. tax updates. accumulations and balances.  Worker spend management: reporting.  Customer accounts: invoicing. expenses.In addition to its human capital management modules described earlier. reporting. Volatility is here to stay. reimbursements. onboard contingents. accounts receivable. This calls for skillful use of cloud 39 . earnings/deductions.  Procurement: information policies for acquiring goods. Workday also offers some functionality related to financial tasks such as:  Payroll: pay groups. Chapter 2 MOVING TO THE CLOUD A BUSINESS STRATEGY BASED ON AGILITY Success will go to those companies that combine business agility and cloud computing to continuously explore new opportunities. Companies that create a foundational business strategy based on agility will do best in this real-time global economy we now live in. Companies that emphasize continuous exploration of potential opportunities followed up by rapid expansion into new markets. It can provide reporting functionality for all of the modules and includes logic to enforce all legal and company-specific accounting rules. services and rights contract labor.  Supplier accounts: all information for vendor management and accounts payable  Cash management: bank statement reconciliation.

social media and consumer technology devices such as smartphones and tablet computers. "Differentiation" (creating uniquely desirable products and services) and "Focus" (offering a specialized service in a niche market).computing and related technologies such as software-as-a-service (SaaS). Figure 2-1 Generic Strategy AGILITY 40 . Porter called the generic strategies "Cost Leadership". He then subdivided the Focus strategy into two parts: "Cost Focus" and "Differentiation Focus". GENERIC STRATEGY Many business strategists use Michael Porter‘s classification of three general approaches to strategy (Porter. All outside factors that may affect an organization make up the external environment. 1998). These are shown below. We should consider the external and internal environment to know more why those strategies are applied.

STRATEGIC REALIGNMENT 41 . We refer to these advantages as core competencies when they cannot be easily emulated and are able to contribute toward added value either in terms of product capabilities and quality or else in terms of more efficient production processes. the source of differentiation must be rooted in unique capabilities or it will not be possible to gain and maintain market share. there is a requirement for an accelerated time to market. whether on cost or quality.Conventional wisdom would suggest that cost leadership is generally less lucrative than differentiation and. A consequence of these changes is a need for agility on two levels. is becoming increasingly difficult in a transparent world with widely published information that travels at the speed of light. This might mean investment in research and development but it will also imply a vehicle for allocating resources amongst product lines and businesses. only permits one clear winner while there may be several successful differentiated offerings. CORE COMPETENCIES Regardless of the generic strategy. The sooner any advantage is implemented the longer the opportunity to skim high profits. If the advantage is to be sustained then it also needs to be an integral part of the organization. However sustainable differentiation. Since first-mover advantage may be short-lived if the competition is able to successfully copy it. obviously. PRODUCT FOCUS Once the sources of differentiation have been distilled into a set of core competencies the challenge will be to determine how to leverage the advantages by developing and promoting the right products. On-demand computing and services can add value by expediting the process of launching new market offerings.

It may be worthwhile to begin with a high-level workshop and follow up with interviews of all key players. In order to maximize chances of success it is vital to perform a multi-tier analysis. SWOT ANALYSIS 42 . so it is necessary to solicit input from all businesses at both executive and operational levels. the next step is to match the two and draw your own conclusions as to how you can best meet and exceed your strategic objectives. Similarly. If you can identify the pain points in current operations and the obstacles that inhibit performance there may well be areas where aspects of cloud computing. often expressed as ratios that can be used as baselines for costs. industry benchmarks. but cloud computing will impact the business across the hierarchy. performance and quality can uncover areas where improvement is necessary in order to address competitive disadvantages or leverage advantages to their maximum. The mechanics will vary from organization to organization.After you have looked at both the developing environment and the internal structure of the company in light of cloud computing. A gap analysis can be a good starting point. service orientation and outsourcing can contribute to realignment with the strategy.

FOCUS ON CORE COMPETENCIES 43 .Figure 2-2 SWOT MAXIMIZE AGILITY Similarly. An agile infrastructure makes it much easier to accommodate innovation and experimentation. suppliers and customers. IMPROVE INTEGRATION A global service-oriented architecture presents many opportunities for improved integration between partners. flexibility is very important to companies in a rapidly changing environment. It also facilitates a quick path to market if pilots and prototypes prove successful. External resources can be easily shared and Web services can facilitate extensive collaboration.

If similar commoditized services are available on the market for a much cheaper price then it is certainly legitimate to investigate whether it would be possible to leverage them. After establishing core competencies and determining whether they lead to a business benefit the key question is whether they are indeed unique and the uniqueness is sustainable. The question is whether it will be possible to extricate the generic function and source it from another provider. A general purpose IT application with customized enhancements for key business processes might fall in this category. 44 . it is worth asking how much of IT should be delivered by internal sources. There may also be alternate forms of delivering the services. There is no universal answer but it is a point that deserves consideration where it applies. The resources released would then also be free to focus on other areas in which they can increase shareholder value.The increased service orientation of cloud computing elevates the importance of identifying and analyzing competitive differentiation. In some cases. you may want to consider the business model for packaging the functionality and selling it as a service to other customers. In the same way that intellectual property can be used internally and/or licensed you may decide that you can have your cake and eat it too. But that doesn‘t mean that they cannot also be leveraged for other purposes. self-service portals can reduce human involvement and lower the costs of the same basic service.based solutions. For example. A related question is whether it is possible to monetize the core competencies in other ways than they were originally intended. The competitive advantages may be intended for streamlining production processes or developing higher quality products. elements of a competency may be truly unique and valuable but they are intertwined with other tasks and responsibilities that are undifferentiated. Again using IT as an example. Particularly if you have software.

The direct impact of utility models is not limited to technology. The markets may also be in flux as new segments gain. Information. And. It is a continuous process that every company should be monitoring. in importance or as some geographical boundaries dissolve while other are erected. IT) become less horizontal in focus it might make more sense to allocate them to the business units or to move the entire organization to a divisional rather than functional structure. There may be a direct effect of on-demand services in some industries.Note that there may be far-reaching ramifications to a refocus of competencies. But even where the effects are not as obvious it may accelerate changes in the landscape so that vigilance becomes much more critical. consulting and business processes may also be made available on demand. On a practical level. SHIFT TO MAINTAIN COMPETITIVE ADVANTAGE Market shift is nothing new. Almost all industries consume these and most of them also provide them – or at least have them potential to provide them. The indirect impact is even more extensive. and other lose. this means that it may be necessary to review and revise strategies more frequently in the future than in the past. The corporate strategy should always reflect changes in the industry landscape not only of competitors but also of suppliers and customers. There could even be further implications. regardless of industry. Even those organizations that do not offer any form of service will undoubtedly engage in transactions with others that do.g. The impact of cloud computing on these shifts can come in two variants. for those companies that 45 . For example. If cloud computing is truly disruptive and impacts this ecosystem then it will accelerate the changes in the partner landscape. If the product portfolio is overhauled there may be cause for reshuffling the resources and responsibilities between divisions. if central functions (e.

They can better align those costs with actual business needs because costs will fluctuate with usage. for example through micro-segmentation or focus on the ―long tail‖ of low-volume merchandises? USING THE CLOUD FOR BUSINESS ADVANTAGE Companies are now at the point where they need to move on from an internal focus directed at maximizing use of ICT resources to an external focus on supporting collaboration and new product development through use of cloud computing. CRM. and a growing array of industry specific applications. This is happening because cloud and SaaS and social media vendors are becoming more and more like utilities offering reliable computing power and basic applications like e-mail. Because of this. companies will outsource more and more of their basic ICT operations so as to better manage their costs for these services. Costs will rise when usage rises but 46 . video conferencing. it may be wise to begin considering alternatives. ERP. If entry barriers are disappearing is there any way to erect new barriers by focusing developments in a direction that is less accessible? If some product lines are under threat are there others that were previously dismissed due to prohibitive investment requirements but are now viable? If some market segments are approaching saturation is it possible to pursue others. Companies are moving toward externally focused web-oriented architecture projects where they begin using SaaS applications and social media and combining them with selected internal application systems to support collaboration with other companies and attract new customers to drive their growth. Over the coming months and years these vendors will develop economies of scale and expertise that enables them to offer their services at a much lower costs than what most companies would spend to deliver those services internally.see changes on the horizon.

can be paid for by operating revenue and costs will drop when usage drops and there will be
no need to renegotiate contracts or continue to pay for large technology investments when
they are not needed. This will in turn enable companies to shift more of their time and
attention to doing things with ICT that add value to their products and provide meaningful
differentiation in the eyes of their customers.
Adoption of cloud computing and mobile consumer devices are causing changes in the
structure and operating priorities of internal ICT groups that are perhaps even more profound
than those changes caused by the introduction of PCs and the Internet.
Cloud computing offers significant advantages in its low startup costs and quick delivery of
computing resources and its pay-as-you-go cost structure. In addition it offers ease of
management and scalability of systems as needs grow. Cloud-based systems are also device
and location independent so people can accesses these systems from anywhere using many
different devices from PCs to tablet computers and smart phones like Blackberry, Android or
iPhone. And finally, cloud computing enables rapid systems innovation in companies to
respond to evolving markets and customer desires.
Some of the primary benefits of cloud computing derive from improvements in cost, risk,
security, flexibility, quality and focus. Let‘s look at each of them.
COST
The most apparent benefits of cloud computing are around cost. In order to quantify the
benefit you will need to perform a complete financial analysis of both a cloud model and any
alternative options. We will pursue this track in more detail in Chapter 9. At this stage, we
will focus on the main cost differentiators.
There can be a significant reduction in up-front investment since there is no need to purchase
extensive hardware infrastructure or software licenses. Instead you can align your costs to

47

actual usage which means that you can allocate costs to the contributing revenue much more
easily and accurately.

Figure 2-3 Fixed Capacity Utilization Curve

You also no longer need to over-provision resources in order to meet spikes in demand. In the
cloud, you do not pay for idle capacity which further reduces costs.
And finally, some benefits which the providers have acquired in terms of economies of scale
and their place on the experience curve will translate into cost savings for the customer.
Certainly, the providers will try to retain most of their advantage as profit but, in a
competitive environment with other providers, you can also expect some savings to be passed
on to customers.
RISK

48

Cloud computing can offload some risks from the customer to the service provider. By
contractually stipulating data protection and disaster recovery provisions, and attaching them
to indemnities in the case of failures, the company can mitigate its own risks.
It also reduces the likelihood of under-provisioning. Since it is not possible to accurately
predict customer demand there is always the possibility that there will be sudden
unanticipated spikes of resource utilization. If the company owns its own resources then there
are limits to the amount of idle capacity which they will procure on the off-chance of a
sudden increase in activity. On the other hand, the elastic capacity of a cloud provider should
not often be exceeded.
There may also be an impact on financial performance through interruptions in billing or
investment activities. Revenue and cash flow recognition may be delayed and distort the
financial picture and there are risks of lost discounts from accounts payable, which can also
damage the credit rating. If that isn‘t enough, then consider the contractual payment
obligations to temporary employees, schedules for equipment renewal, overtime costs,
shipping costs and travel expenses, which can all be adversely impacted.
It is impossible to completely remove the threat of departmental cloud use. However, if the
functionality is available on an authorized and supported basis then the incentive for
unauthorized and unmonitored usage declines.
SECURITY
Security is usually portrayed as a challenge for cloud computing, and rightfully so.
Nonetheless there are several benefits that cloud computing may offer with respect to
security. That is not to say that these benefits are necessarily exclusive to cloud computing,
merely that they align very well with cloud computing.
Cloud providers typically undergo very strict security audits. An enterprise may also institute
the same audits but, on average, many businesses do not enforce the same level of rigor as a

49

cloud provider. On a similar line, the cloud providers have access to the best-of-breed
security solutions and procedures. They have also been forced to inculcate a deep sense of
security concern in their administrative staff. Again, this is not typically matched by smaller
organizations.
The cloud also offers a platform for many security functions ranging from disaster recovery
to monitoring, forensic readiness and password assurance or security testing. Its location
makes it ideal for centralized monitoring.
It is easier to isolate customer and employee data if they are managed in different
environments. It might therefore increase security to segregate the data such that customer
information is housed in the cloud while employee information is processed on the internal
network.
Virtualization carries with it the inherent advantage that it is much easier to deploy
preconfigured builds. It is possible to pre-harden these by locking down all traffic,
eliminating unnecessary applications and features and applying the latest security patches
Finally, there is some arguable advantage to the fact that the cloud obfuscates the physical
infrastructure. Since virtual images may be brought up anywhere in the cloud and tend to
move frequently it makes it much more difficult for a hacker to launch a topology-based
attack.
FLEXIBILITY
A cloud infrastructure adds considerable flexibility and agility to an enterprise architecture. It
makes it much easier to roll out new services as they become necessary and to retire old
applications when they are no longer needed. There is no need to procure hardware for the
former or to cost-effectively dispose of the equipment in the case of the latter.
Similarly, a particular service can scale up and down as needed. There are cases where
resource demand has spiked ten-fold overnight only to fall back to its original level shortly

50

The elasticity of a cloud allows the enterprise to exactly match the resources to the demand without overpaying for excess capacity or losing an opportunity to address market demand. The usual lead time for procuring necessary equipment can be compressed to a few minutes when resources can be provisioned on demand. a globally replicated cloud facilitates access from any place using any device at any time and therefore contributes to user flexibility and productivity. experts can refocus on activities and opportunities that help to solidify the core competencies of the firm. Once service evolution is automated. QUALITY Quality of service in all dimensions is a major concern around cloud computing. And finally.after ward . Cloud service providers have great economy of scale and specialization. These responsibilities extend from user provisioning and support to application management and troubleshooting. They have developed rigorous processes and procedures to maximize uptime and optimize performance. The flexibility also facilitates a faster time to market. Ultimately the speed and reduced commitment also lower barriers to innovation. FOCUS The fact that some of the IT services are outsourced to a cloud provider reduces the effort and administration that is required by the corporate IT department. which can encourage a more agile organizational culture. 51 . They run best-in-breed software to monitor and manage the infrastructure and they employ some of the most skilled practitioners to oversee the management tools. But in many cases it is actually a benefit.

compromise the information. which lock the customer into a given vendor‘s service. in fairness. FIANCIAL STRUCTURE The absence of capital investment and reduced fixed costs may seem very attractive but there will probably also be significant implementation and migration costs associated with the move to a cloud platform. A common obstacle to the adoptions of cloud computing is the fact that the service provider hosts sensitive data – potentially in a multi-tenant environment. a BT survey in November 2009 found that the majority of CIOs do not see value in cloud computing (BT Global Services. 2009). Some of their concerns can be dismissed as over-generalization of bad experiences from previous hype cycles and lack of familiarity with the essence and benefits of cloud computing. That doesn‘t necessarily mean the service cannot be moved to a cloud but double-payment to multiple providers is certainly a significant financial barrier. Sunk costs in existing infrastructure may also act as a deterrent toward migration. However. The customer must consider the host to be trustworthy enough not to intentionally. Another challenge can be the existence of current contracts with service providers. The fact that there is only limited standardization of cloud functionality leads to interoperability barriers. there are also legitimate drawbacks and challenges that new technologies bring with them.BUSINESS APPLICATIONS WITH THE GREATEST POTENTIAL Although there are some very compelling benefits to cloud computing. SECURITY AND RISK The security and risk concerns are probably the best known and most challenging to address. or inadvertently. These may preclude (or heavily penalize) early exits. They may represent an emotional barrier but it is much more important to examine the future costs. This presents 52 .

It can be difficult to manage the encryption keys needed to protect all channels of communication. which can complicate operational processes such as incident management. ORGANIZATION Shifting functionality to the cloud essentially means outsourcing that service. However. This evaluation is an additional burden. Again. there are many questions about support in a multivendor. There may be uncontrollable sources of latency or data transfer bottlenecks. They have only limited recourse if the system performs unreliably or doesn‘t scale to their required capacity. INTERGRATION AND TRANSITION There are numerous technical challenges that an enterprise needs to address if it is considering adopting a cloud model. And across these. It does not necessarily mean that the groups who were performing that service are no longer needed. there are security precautions to consider but there are also more general problems of interoperability and standardization of interfaces. 53 . And outsourcing usually has personnel implications. A general governance problem can give many customers cause for concern. it does mean that there will be a need to reassess their responsibilities and ensure that they can continue to add net value to the business. It is not trivial to integrate the networks in a reliable. which may need to connect between organizational boundaries.risk if the vendor faces insolvency or if the customer subsequently chooses to switch vendors. multi-provider environment. which can slow down the adoption of a new technology. scalable and secure fashion. It can be difficult to maintain data integrity when critical parts of typically synchronous activities need to be handled asynchronously. It is also a challenge to integrate applications.

In a typical scenario. The relevant impact on revenue and cost of a given proposal must be quantified to take the difference in order to determine if it is profitable. which temporarily increases the costs above the current baseline. the ongoing costs reduce to below the baseline. 54 . the profit equals the revenue less the costs.CONTEMPORARY WAYS BUSINESS ARE USING CLOUD RETURN ON INVESMENT In simple terms. After this point the investment can be considered profitable. The financial budgeting task is to determine whether the total is better or worse than if the investment is not undertaken. Assuming an initial increase and subsequent reduction in costs the curves should intersect when the ―payback period‖ is complete. Figure 2-4 Cumulative cost of investment An alternate visualization would be to look at the cumulative cost of the investment over time (Figure 2-4). if it is a one-time investment (such as the transition to a cloud service). However. after the initial investment.

Until then. a profit is made in excess of what could be achieved from alternate investments. the biggest challenge is that a great portion of these costs have already been incurred and are therefore not relevant for near-term future considerations. Similarly. RELEVANT COSTS Capital costs: The expected implication of utility computing is that should be seen a curtailment in capital costs. Inclusion of optional future costs can only be justified in the equation. However. If it can be terminated. But for financial planning purposes these are independent. It may be necessary to migrate other users or applications in the future. once the technology has reached the end of its expected refresh cycle then it is valid to include the costs in the financial decision. Firstly. they should be ignored. Secondly. non-recurring costs. 55 . However. the costs become relevant. or in rare cases. Transition costs are truly one-time expenditures. to roll back and re-apply a migration. however. the relevant costs must be identified and measured. It may also entail penalties for early termination of existing contracts and/or double-charges that must be sustained in order to avoid violating the contracts. the existing equipment may be still operational. Over the long run there should definitely be a reduction in some of these areas. it is not sufficient to make a profit. Transition costs: would include migration of data and applications to the selected cloud platforms and any necessary retraining of employees. since relevance is directly related to the duration and renewal terms of the lease. Both property and equipment leases are common as an alternative to purchase. include data center property and equipment costs as well as computer and networking infrastructure costs. Secondly. This only superficially changes the question.There are two complicating factors.

Reductions in user productivity and satisfaction have direct cost implications but there is no standard way to measure them or to value them. the operational expenditures now include not only the administrative tasks but the software licensing and equipment lease. customer. If there are directly attributable redundancies then the cost of the severance packages is a relevant cost. While these soft costs are all valid. they typically meet with significant skepticism from financial and executive officers. There will be some benefits that the cloud service provider can pass on to the customer in terms of economies of scale and advanced position on the experience curve. Employee satisfaction leads to reduced turnover as well as retention of company. Personnel costs: if the transition to the cloud impacts employee roles and responsibilities. and particularly if it involves significant outsourcing. User productivity can be affected by the up-time of systems. as a result of a transition of service increase or decrease in the number of calls/user/month or the average resolution time. However. they are also difficult to estimate both in terms of magnitude and frequency. quality and efficiency of support. Underprovisioning costs can have huge impact. On the other hand. then it is necessary to look at the financial impact incurred. the reliability of applications and connections. 56 . Customer satisfaction levels are tightly correlated to customer retention (which is generally considered to be vastly cheaper than targeting new customers). product knowledge and lower recruitment costs). Support costs: these may change if. Soft costs: There are many internal metrics that do not directly equate to financial gain or loss. The premises of the calculations are easy to challenge and the estimates cannot be empirically substantiated.Operational costs: The operational costs are an interesting area to examine.

The most likely tangible revenue benefit is typically associated with any online services that the organization offers of its own. network connectivity.In order to make a viable proposal for cloud computing it is necessary to consider all relevant costs of the baseline system and compare those with the costs of a cloud-based option. If the sum of the cash flows is greater than zero then net present value is positive. It reflects the time value of money. which directly equates to increased profitability. In some cases. there will be no reason to expect a change in costs. Another common approach for comparing investments is to calculate the internal rate of return (IRR). Cloud services will very often be billed according to usage rather than footprint so it is critical to devise a means of calculating equivalence. contract management. This lack of insight may convince some business leaders to imply a high rate of risk and therefore demand an exorbitant rate of return. The cost-basis could also differ. client hardware. It is difficult to obtain empirical data on the possible returns for cloud computing and without historical data you cannot calculate a standard deviation. Some of the line items to consider include: Server hardware. storage hardware. Capital budgeting: is a set of tools that can be used to measure the opportunity cost of an investment or to compare multiple alternate investments. service management. It may be advisable to be conservative in estimating cloud computing until the 57 . it is also feasible that new applications will imply new end-user access devices. operations. software licenses. The elasticity of resources allows the company to react more quickly to surging demand. network infrastructure. The IRR is the rate that would need to be selected in order for the investment to break even (with NPV=0). However. security infrastructure. To keep it simple. The costs of each of these items may not easily be accessible and the comparison between historical costs and cloud-based estimates may not be straightforward. support. assume each cash flow is at the end of a fixed period and that each period has the same anticipated rate of return (i).

The bluntest instrument. it is always in your best interests to determine the risks and variability as precisely as you can so that you can address and monitor areas of uncertainty. However. is not to use any: just apply a standard companywide hurdle rate. faster billing and collection as well as deferred vendor payments. Other means of improving cash flow include reduction in inventory levels. which can be made available to customers who run into these obstacles. And if they diverge then you may want to investigate the reasons. the same challenge of an initial expenditure for implementation and migration also impacts the cash flow. but the shape of the curve will change. optionally padded with a very subjective safety buffer. operations and investments. If it is possible to reduce capital expenditures it will positively affect both the depreciation schedule and debt financing. Cash flow: represents the amount of cash received less the cash outlays from financing. Cloud computing doesn‘t directly impact any 58 . identify similar projects. This is certainly easiest but if you do have some additional insight you can bring to the table it is well worth considering. The same benefits of cloud computing typically apply for cash flow as for profit. which is a fancy name for borrowing. Some IT organizations have access to financing. Even though a project may have a compelling long-term benefit.services mature. observe market. but probably also the most common. If all indications converge then you can move forward with increased confidence. There are some options but they have their limitations: poll experts. particularly in a very conservative economic climate. The benefit to the customer is that it is possible to accelerate the payback period by deferring some of the costs. However. Assessing the financial risk of an emerging technology is challenging. the fact that it requires an investment can be a deterrent. It is possible to circumvent the obstacle with ―financial engineering‖. The total amount will still be the same or actually larger due to the financing charges.

The ROA measures how well a company can leverage its assets and is often used to gauge efficiency. in order to be compared favorably to other companies in the same industry. Imagine that a hypothetical example of 90% of assets being in IT then ROA tenfold can be multiple by moving to outsourced services. Public cloud services can help in this regard. Since it varies inversely according to the asset valuation. Even if the resources are shared within the organization it may be possible to leverage the fine-grained instrumentation of the provider in order to apportion and assign the costs to the appropriate owners. which is much more attainable when data and processes are standardized and securely externalized – features that are very closely associated with cloud services. 59 . A high ROA can lead to better analyst and shareholder valuations and can even improve the cost of financing debt as it may impact the credit rating. so the efficiency will truly improve that dramatically. As they are not owned by the customer they do not appear in the balance sheet and therefore imply lower asset values than privately run services. they all require very tight integration and flexible terms with partners along the supply chain. Financial visibility: A utility pricing model greatly enhances financial visibility. Return on Assets (ROA): computed as the net income divided by the average total assets. Since costs are directly correlated to the revenue-generating products. there is less need for assigning overhead costs and differentiating between direct and indirect costs. However. Note that the ratio becomes dubious if taken to extremes. If the resources are allocated to product lines during the subscription process then it is possible to apply an activity-based costing model which assigns the cost directly to the product lines that consume them.of these. it is in the company‘s best interest to keep the assets to a minimum. and services that incur them. For this reason.

0 services that are almost exclusively delivered over the Internet. There are a growing number of Web 2.TIME GLOBAL COLLABORATION COLLABORATION Since collaboration involves establishing connectivity between people it is natural to also use a technology that is built on networking and utilizes a common infrastructure.ROA is not generally used to compare companies in different industries and should never be considered in isolation. GENERATION OF COLLABORATION Figure 3-1 Citations per article versus number of collaborating countries Google Apps 60 . But even some of the more traditional applications such as desktop productivity and conferencing can benefit from cloud computing. Chapter 3 GLOBAL IMPLICATIONS OF THE CLOUD REAL.

These include: Gmail.Google offers a set of applications under the banner of Google Apps both as a free service to individuals and a competitively priced service for companies or individuals who may require higher storage limits.3.  Google Calendar is a time-management web application. and Sites:  Gmail (also called Google Mail in some countries) is a webmail service provided with a thread-structured. the Ajax-driven interface allows users to drag-and-drop events from one date to another without reloading the page. search-oriented interface. Google Calendar. and IMAPv4. In addition to browser access it also supports POP3. The user experience is characterized by the extensive use of Asynch ronus Javascript and XML (AJAX) to pre-fetch and pre-send information without the user needing to refresh thereby dramatically reducing response time and improving the user experience. Talk. In addition to viewing. Users can add calendar events by typing informal English phrases. Gmail is noted for its spam filtering which leverages a community-driven system: when any user marks an email as spam. the system leverages the information to identify similar future messages for all Gmail users. such as "Team call at noon next Sunday". monthly. and agenda. It supports view modes such as weekly. Docs. They can be shared either with read-only permission or full control and either with a specified list of people only or made public to everyone 61 . adding and modifying entries. Gmail runs on Google Servlet Engine and Linux-based Google GFE/1. They can also set the number of days to show in their custom view mode and can add descriptions to all events in the Google Calendar. Other features of Google Calendar include offline support and to-do lists. At the back end. Commercial customers can also use Postini as an alternative to Google‘s own malware detection. Multiple calendars can be created and shown in the same view.

By default. users can be notified of changes to any specified regions via e-mail Typical of windows Windows Live Family Safety (parental control system providing website blocking. PDF. forms and presentations can be created within the application itself. It allows users to create and edit documents online while collaborating in real-time with other users. RTF. Windows Live Mail (Email client for Windows Live Hotmail. content filtering and web activity monitoring).  Google Docs is a web-based desktop application suite. spreadsheets. They can also be saved to the user's computer in a variety of formats (the ISO standard. Windows Live Movie Maker (Video editing software). Yahoo! Mail and any POP accounts) Windows Live Messenger (Instant Messaging client). imported through the web interface. Google Docs serves as a collaborative tool for editing amongst users and non. Windows Live Sync (File synchronization between user desktops and other computers).users in real time. Shared documents can be opened and edited by multiple users at the same time. Documents. Microsoft Word). Google Talk servers uses (XMPP). Windows Live Writer (Desktop blog publishing 62 . which accommodate peer-to-peer signaling for multimedia (voice and video) interactions. Windows Live Toolbar (Browser toolbar including Desktop and Internet search). In the case of spreadsheets. Google Talk (GTalk) messaging and internet. Gmail. Text. an extension to XMPP proposed by Google. including offline messaging and voice mailing. Open Document format as well as Open Office. Windows Live Photo Gallery (photo management and sharing). or sent via email. Instant messaging between clients and the standard Extensible Messaging and Presence Protocol real-time are extensible messaging and presence events. they are saved to the Google servers. VoIP (Voice over IP) is based on the Jingle protocol. HTML. which provides is a Windows and web-based application for instant telephony.

The main components of the BPOS suite include: SharePoint Online offers content management with version control and support for wikis. Excel and PowerPoint so that the desktop version of the software can easily retrieve workspace files and store modified versions back online. web site design and hosting services as well as a contact manager to organize customer information including contact history and sales information. It includes domain name registration and business email accounts. and applications. surveys and RSS feeds. The standard offering operates using a multitenant model and comes at a very attractive price point for smaller organizations. It also supports Forefront anti-virus. The dedicated option involves a higher degree of isolation. Users can share documents. Office Live Small Business is an externally facing service for small businesses. passing control to anyone in the meeting. Office Communications Online provides presence capability and messaging. audio broadcast or 63 . The workspace also provides tabular lists with templates for Tasks. Users can share single documents or workspaces with multiple documents and can collaborate as a group. Microsoft Silverlight (Web application framework including video streaming plugin) Microsoft Office Live offers two services for small businesses: Office Live Workspace is an internal service for storing and sharing documents online. They can also include video from multiple webcams and use integrated VoIP. Office Live Update offers integration from Word. blogs. presentations.application) and Microsoft Office Outlook Connector (Outlook connector to Windows Live Hotmail and Office Live). Events and Contacts. Microsoft Office Live Add-in. which can be synchronized with Outlook. Office Live Meeting offers both live and recorded whiteboard and presentation sharing capabilities that can include audio and video. which is often required by larger enterprises. It is effective for showing demo applications and conducting web tours. The Business Productivity Online Suite (BPOS) caters to enterprises and comes in two packages called Standard and Dedicated.

The meeting history facilitates billing and can provide information on attendance and test results. Regardless of how you categorize them they are an important emerging area that needs to be considered. These solutions could equally well fall into the category of Storage as a Service since most of the computation takes place in an on-premise client stub. edit. On the other hand. Mozy Backup 64 . Mozy is more typical of a personal backup solution. Some of the offerings on the market include Box. and play back meetings after the event for review and compliance purposes. NEW WAY OF WORKING BACKUP AND RECOVERY Cloud based backup has an intrinsic benefit that the physical storage is generally very far removed from the users and their source data.dial-in conferencing as well as hybrid combinations. It is possible to record. the end-user interaction is more typical of Software as a Service. As such it is resilient to many of the risks of natural disasters and other localized outages.

such as Outlook PST files. It includes support for open and locked files. through the web. At the beginning of the backup process. The tool backs up common business applications such as SQL and Exchange. all files are locally encrypted with 448bit Blowfish encryption.Figure 3-2 Intensive load of backup and some other Mozy allows users to specify folders and files that should be backed up to the cloud on a regular basis. Users and administrators can restore data via the Mozy software client. Mozy only backs up new or changed portions of files in order to optimize network traffic. Users can also specify how much bandwidth they want to dedicate to their backups so other high-priority services can run unimpeded. Scheduled backups can be performed daily or weekly at a specified time of day. Windows users can also restore data via right-click or through the Mozy Virtual Drive. The service automatically detects and backs up new and changed files while the computer is not in use. or by ordering a DVD restore. The encrypted files are then sent via 128–bit SSL to a managed data center where they remain in their encrypted state. The professional version of Mozy adds the role of administrators who can view individual and 65 .

its ZCloud API is publicly documented and can be leveraged by other storage services to become potential destinations of the evergrowing eco-system of backup-to-cloud applications. They can also create customized email alerts regarding the backup health of their account. The central management server can interact with application agents that cover Microsoft Exchange. It is of interest for enterprises that may need to run in a hybrid environment. It is based on the Amanda open-source backup software and provides a central console where administrators can specify hosts. taking into account needs such as 66 . such as: Sentri7 is a real-time patient surveillance system that can provide a list of hospital patients who are most at-risk from medication-related falls so that appropriate precautions can be put in place. Zmanda Zmanda is a backup solution that takes a very different approach. labor cost. It aids compliance and provides visibility into time off. paths and backup destination. Zmanda does not run in a cloud. Active Directory and SQL Server as well as Oracle 10g and Oracle 11g. However it also provides an option to back the data up to Amazon S3. In particular. its employees. changing and communicating staff schedules.aggregate account history. and other reports generated for individuals. Healthcare Pharmacy One offers several services in the healthcare sector. The builtin rotations and templates include set shifts and cycles. reset account password and distribute license keys. INDUSTRY SOLUTIONS One of the most intriguing areas for cloud services will be in delivering vertical solutions that help to automate processes in particular industries. Each customer subscribes to a license of ScheduleRx that is unique to that department. shifts and departments. shifts and other requirements. ScheduleRx is an online pharmacy staff scheduling solution that streamlines creating. It is a centrally administered solution that can backup data to disk and tape. It does not use any proprietary data formats so it is relatively easy to integrate the solution with other storage solutions.

Quantify is a customizable clinical documentation and reporting tool for hospital pharmacies and health systems. Some of its functionality includes: Data Management brings together the supply chain community of retailers. carrier notes. ground express truckers and dedicated air charters. Pharmacy managers can run detailed reports trends in intervention acceptance. It also manages the end-to-end shipment process with functionality to administer the shipment. medication errors. Procurement integrates demand signals across each node to facilitate control over the complete purchasing process. transportation expense. time spent doing clinical activities. and adverse drug reactions via the web or handheld computer. suppliers and distribution. Pharmacists document and monitor their clinical interventions. Logistics One Network offers solutions in the retail and consumer products sector and supplies technology for logistics service providers.double shifts. night/day shifts and weekend rotations. track cargo location. detailed logistical visibility to shippers. training. suppliers and carriers by providing them with a data sharing framework. Transportation Active On-Demand's APT is an Internet-based transportation management system (TMS) that connects commercial shippers with hand-delivered couriers. customers. preferred consecutive scheduling patterns. consignees. and costs avoided. Manufacturing is an end-to-end process flow for manufacturers from production to 67 . The Carrier Dashboard provides a view into carrier activity over a date range and shows shipment volumes. and carrier quality ratings with drill. matching requests to quotes. and carriers.down functionality for data mining including cargo records. and provide complete. Order Management delivers an automated order process that can be viewed and collaborated between sales representatives. and a complete audit trail for a specific shipment.

how much to stock and where to source. 68 . Store Operations aligns all the processes around store shelf. Replenishment defines replenishment policies ranging from static to completely dynamic as well as postponement techniques to generate the most accurate orders. It caters to manufacturing processes ranging from make-to-order to custom and mixed mode. sense and respond transportation and appointment scheduling solution that integrates into a company‘s overall supply chain process. 2004) the Supervisor of Midwives must be informed and consulted with regard to the requisite preparation and experience. procedures and framework outlined in this document apply. 2004) and many aspects of practice are laid down in statute. Inventory Management offers inventory visibility and management across multi-party users. It works with the underlying replenishment engines to maintain optimal inventory levels throughout the value network across stores and plants. Spare Parts is a module to manage and replenish spare parts. multi-party. the principles. Demand Driven Logistics is a multi-tenant.purchasing to inventory. Rule 6 details the responsibilities and scope of midwifery practice. Deployment coordinates activities of deployment planning and execution to meet customer‘s service levels with the lowest total landing cost. using analytics that not only provide visibility into historical issues but also project future problems. It includes decisions on where to stock. continuously adjusting forecast according to the latest demand patterns. As stated in the ―Midwives Rules and Standards‖ (NMC. If midwives wish to adjust or add to their scope of midwifery practice. Demand Management integrates a retailer and its suppliers in processes focused around actual shelf consumption with a collaborative architecture. MIDWIVES AND MIDWIFERY PRACTICE The role of the midwife is clearly defined in the ―Midwives Rules and Standards (NMC.

we believe personalized and proactive insights are required to cut through the noise. we want to show you what we‘re doing with them and how we‘re harnessing these powerful forces to transform work. we have an incredible ability to achieve—but only if we can focus. It‘s about how we build relationships. brands can be made or broken with a simple Tweet. We recognize that one size doesn‘t fit all. we believe the future of work is all about working like a network. and we find our next job through contacts in LinkedIn.But this year. with information coming from the top down or held up in unconnected 69 . Meanwhile. and that companies need customized solutions that can be developed in weeks. EMPOWERING PEOPLE AND ORGANIZATIONS Today. The innovation you‘ll see from us here today—and in fact over the next few years—is based on four core beliefs:  First. and respond to ever-changing conditions. And we need technology to help us. so that we can focus our energies on accomplishing big things. share information. the world is a giant network with people hyper-connected to information and each other.  Third. As humans. This year. We talk to our friends and family using Facebook. we want to do more than just talk about technology trends. we believe in the power of an open development platform and a robust ecosystem.  Second. And it‘s founded on an openness and transparency that will drive a new level of productivity.  And finally. and if businesses have the tools they need to responsibly manage security and compliance. many organizations continue to work the way they always have: they operate in hierarchies. and we‘re able to accomplish more together because we are so connected. not years. Information and knowledge move faster. we believe that enabling this future of work is possible only if people are empowered to work anywhere and on any device.

We believe this is the start of something game-changing—building digital memory across applications to create a highly personalized experience that helps people get more done. Outlook. people. We know that not everything is best expressed in documents or conversations. we can intelligently connect and surface the most relevant content using the Office Graph. cloud. big data. To be more responsive. One of the first experiences to leverage the Office Graph is code-named Oslo. Many organizations have created dynamic portals. leverage multimedia. This is the power of social. We have of lot of work going on in this area that will show up this year. that serve unmet needs in most organizations today. Office. conversations. and that allow organizations to work more like a network. using a network of highly informed and highly connected people. documents and more. deliver more relevant information. groups. We believe next-generation portals should be simple to create and customize in the cloud. SharePoint. The Oslo application surfaces what might be most relevant for each individual based on what they‘re already doing in Exchange. and—most importantly—be accessible anywhere and on any device. and Yammer. and mobile apps to surface what matters most to employees. Today we‘re showing examples of innovation in the cloud that I‘m really excited about. By leveraging insights from Office 365 and mapping the relationships between people. Office Graph will power many new personalized and contextual experiences for people using Office 365.systems and broken communication channels. have high engagement via social interactions. files and conversations through machine learning. but today we‘re showing a new 70 . We need to flatten the information flow so that we can work better together and serve customers better. Oslo serves up in a single user interface personalized insights based on meetings. Here we‘re also working on some bold new approaches. web sites. Lync. and machine learning coming together. we need to evolve.

and that will also help our customers see more value by having direct access within Office 365 to the things they want to do every day. as a ready-made solution. we want to give customers maximum agility to adapt our products to specific business needs. or not so well. But it is hard to apply conceptual advice to thousands of applications which many enterprises have in their inventory. A faster. First. Second. Yammer. and Azure Media Services. we want to bring partners a great opportunity to grow their business.social video portal built on SharePoint. It‘s been very exciting in the last few weeks to have partners ranging from Go Daddy to DocuSign building solutions and services around Office 365 that will help them grow. Suitable workloads might include:  Storage as a service 71 . We have two objectives in our development platform. NEW REALITIES AND NEW OPPORTUNITIES CLOUD OPPORTUNITIES The preceding guidelines were an attempt at some general observations on what constitutes an optimal cloud computing application. albeit somewhat rough. in a distributed and public environment. categorization process is to look at which classes of software may typically lend themselves best to cloud computing.4 million developers—nearly half of whom are SharePoint developers—building on the Office platform and Office 365. our fastest-growing product in history. whether you build or buy these solutions. the Office Graph in Office 365. We see more of this great opportunity ahead for our developer partners. OFFICE 365 FOR DEVELOPERS There are already more than 3. Theoretical advice can be useful for understanding the background as to why particular software works well.

g. Backup/Restore as a Service  Single virtual appliances  Test and pre-production systems  Software development tools  Batch processing with limited security  Mature packaged SaaS offerings (collaboration. the following are not typically suitable:  Sensitive data (e. healthcare records)  Multiple co-dependent services (high throughput online transaction processing)  Applications with stringent audit and accountability requirements  3rd party products without virtualization or cloud-aware licensing  Applications that require detailed utilization measurement and chargeback  Applications that require a high degree of flexibility and customization  Systems that rely on specialized infrastructure. primary LDAP server. CRM. physical location or custom hardware 72 . HR)  Isolated workloads without latency sensitivity and well defined APIs Figure 3-3 On the other hand.

It‘s a good thing for them that many of you reading this are rolling out cloud solutions with those goals in mind. Capacity problems and user dissatisfaction can be an opportunity for change. While this has been successful for many clients I‘ve worked with. BUSINESS OPPORTUNITIES Business leaders. An alternate approach for preselecting candidate applications is to consider systems that currently represent pain points. As a CIO. quite often. Unfortunately. Cloud solutions are attractive to business leaders as they provide an opportunity to quickly deploy new features without following the rigor imposed by traditional IT organizations. traditional IT organizations have spent the past decade or longer integrating large business systems and data to deliver to their business partners a holistic view of the customer. Data models must be modified and coordinated with all users and deployed during scheduled.This is only a rough list. intended as a starting point for finding some suitable applications. but how can I blame them for taking advantage of an opportunity to grow revenue when my own organization is unable to respond fast enough? 73 . must move quickly to drive revenue and profits up or bring costs down. operational or functional terms. the counterweight to that achievement is a process-heavy IT organization unable to rapidly add new features. infrequent releases. it doesn‘t remove the need to diligently assess any candidate application before even beginning to plan for its transition to the cloud. I may not be happy about my business partner excluding me from this new relationship. especially in today‘s economy. Large. And. whether they be in financial. funds have already been allocated for these changes. This limits a business partner‘s ability to react quickly to changing business conditions or rapidly deploy new features to market. Even major upgrades may signal that there are deficiencies with the current system that need to be addressed.

The cost of entering the market is much less for cloud solutions than when we were in a mainframe or even client-server world.com.Figure 3-4 GLOBAL OPPORTUNITIES Widespread distribution of Internet technology to developing countries around the world creates even more opportunities. Sound farfetched? Then let me introduce you to my friends in Syria cranking out high-quality Ruby apps. All can be had via the Internet at a fair. If you‘re sitting there idly dismissing the thought that lessdeveloped countries can compete with you. But isn‘t 74 . it‘s so successful that you hire more staff to meet demand. My point is that cloud services will generate many successful business opportunities in developed countries as well as developing countries.com. You need help building some of your interfaces so you contract with another developer in Egypt on eLance. There is no requirement that small development shops around the world have to purchase expensive hardware or software. Competition from service providers in low-cost developing countries will inevitably increase. Once your service is ready to roll out. You access the Internet to take a course in Ruby on learnable. You build a new service based on Ruby using Heroku. What prevents them from doing the same thing? Imagine you‘re a young developer in a Middle-Eastern country wanting to learn the latest Ruby technologies. equivalent price. remember how quickly you set up your hosting and technical infrastructure via the Internet.

Persistency: The cloud is equipped with a highly reliable backing store. The network traffic and bottlenecks may pose a short-term problem for latency-sensitive thin clients and batterypowered mobile devices. But. large and small. But it simplifies the job of the application developer.even a ―fair price‖ out of reach for shops in low-income locales? Maybe. The cloud lends itself particularly well for new applications which do not have some of the transitional drawbacks of legacy tools and can immediately incorporate design considerations that leverage its benefits. corporate organizations with larger budgets—a key goal for any cloud solution provider. replacement or retirement. think of that small business in Ghana and the advantage they could gain over their competition by accessing low-cost solutions via the cloud. pervasiveness and increased context. Cloud Spring will focus on how enterprises. it is just as important to determine if there are any new strategic opportunities that might lend themselves to a cloud service delivery model. are taking advantage of cloud computing to achieve success. NEW OPPORTUNITIES At no point in the process should you lose sight of the fact that cloud computing has a disruptive impact across multiple layers of business activity. Businesses in developing countries with access to the Internet and modest means can even use the cloud as their competitive advantage. but have you ever heard of micro-financing? On a parallel track. such as persistency. I‘ll also focus on how small businesses are building cloud solutions to gain access to large. It‘s critical to look at the existing applications for potential transformation. 75 . By simply using lowcost CRM solutions. I‘ll look at new cloud services entering the market and innovative ways businesses are using cloud services. they could improve customer service and satisfaction and grow their business faster than their competition.

Both of these trends may facilitate greater adoption of Internet-based services and increase pressure for resolution of geographical obstacles such as disparate legal compliance or latency-sensitive applications. All personally relevant information can be retrieved through a variety of channels. Increased context: Network-centric services are able to build an increasingly complete picture of the user‘s behavioral patterns and immediate context based on timing. This information can be leveraged through analytics to improve the user experience and target user-specific business operations. we should also monitor the potential for political and economic developments to influence cloud computing. the evolution of the global economy from a manufacturing to services base means that a higher percentage of product will be service-based in the future. Chapter 4 FUTURE TRENDS ECONOMIC As described at the beginning of the book. cloud computing may have a dramatic impact on the economy. At the same time. One area to consider is the increasing globalization of markets.Pervasiveness: Synchronization addresses the problem of multiple devices which may not always be connected. 76 . locations and activity. However. There may be new applications that are able to exploit salient characteristics of the cloud to create additional revenue streams and optimize current business processes. The elimination and reduction of regional boundaries in the physical goods and labor markets mean that it is much easier and more commonly accepted to draw products and services from remote sources.

The democratization of the economy is perhaps most visible in the area of printing. the trend is also apparent in software publication with marketplaces that have spawned ecosystems based on a number of platforms from the Apple iPhone to Google. 77 . This requires a broad understanding of both business and technology and the ability to synthesize the two – a skill that is not likely to be common and will be in high demand. with a variety of services that allow easy publication of books. There will be an increased need for unique jobs that combine resources in order to fulfill ad hoc requests and rapidly adapt business processes as requirements change. magazines and photo albums. However.Figure 4-1 A shrinking workforce combined with the ambition of many organizations for increased agility mean that permanent employment numbers are likely to decrease over the coming years. such as web services and crowdsourcing. As a result companies must examine opportunities to automate any repetitive work and integrate a wider variety of productive pools.

or distribution of low-volume items. it has become more generations. especially from the younger generations. CULTURE From a culture perspective. As recently as ten years ago the lack of sufficient information was a common obstacle to achieving business and personal objectives. The absence of economy-of-scale advantages also puts pressure on larger organizations with higher operational overhead to find creative alternatives for fostering a competitive advantage over their smaller competitors. As companies strive to retain their top talent they will need to consider preferred applications and working environments of their employees. Many employees will accept it but some of the brightest and best may choose to work for a competitor instead.The fundamental impact of removing entry barriers to these segments is twofold. Another personal consideration is the increasing challenge of cognitive overload. Simple functions such as operating an electronic appliance might have become impossible if a manual was lost or misplaced. As computing has become a part of life for many people. it has become more difficult (and less desirable) to segregate personal and work activities and tools. by removing the fixed costs that are associated with each item. Forcing someone to work with unfamiliar tools can be perceived as akin to enforcing an unpopular dress-code. The result of this indirect influence is unpredictable but you would expect that it will increase the variety of sources of differentiation. People didn‘t know how to perform a given operation and were uncertain how to obtain the required instructions and inputs. one of the most important trends is that of consumerization. It enables the so-called ―long tail‖. 78 . By increasing the number of product lines it enhances the diversity of the market.

sorting. or may not. such as Second Life. time. reflect the personality and identity of the user. Even. be running on a physical machine. piece of data. At the same time. it is still possible today to have questions for which it is difficult to find the answer. However. online forums and newsgroups provide a means to establish a virtual persona that may. But it will also need to extremely aware of a user‘s context (location. As virtual reality hardware. travel direction. A very rich user experience can also be useful for educational purposes and collaboration in the enterprise. But the more common problem is that. there is too much information that needs to be filtered in order to obtain an authoritative. users are becoming more interested in their virtual experiences as these blend with their physical reality. businesses or 79 . categorization and analytics to help users manage the vast repositories of both structured and unstructured information and to easily find and extract the information that is of interest. or may not. from gloves and chairs to goggles and helmets. or at least reliable. the fact that it is an electronic process means that it is an inherent abstraction of reality and the user‘s experience. VIRTUAL AND PHYSICAL CONVERGENCE Cloud computing is primarily a virtual phenomenon.Of course. Regardless of the actual operating system that may. weather. while the answer is available and accessible. advanced graphic visualization and a plethora of human interface mechanisms. exemplify the fascination that humans can have with a virtual experience. to become more advanced these experiences will only become more compelling. Services. the virtual experience is not limited to entertainment. There is a growing need for filtering. Tomorrow‘s virtual world will need to consider multiplicity of personas.

NEXT-GENERATION NETWORKING Since cloud computing is fundamentally reliant on networking.friends nearby) as well as past actions and experiences in order to guide the user and optimize recommendations. The Internet connection that users can typically afford from their home. and expect from their hotels and office locations. any changes or developments in connectivity will have a direct impact on the services and service quality. and network access points. Persistent conectivity As networks. will dictate the richness of the user experience. wireless and optical transmission mechanisms may provide a means for dramatic increases that not only enable a richer user experience with high-definition video and high-quality audio. weather-forecasts) are already becoming available as services. Broadband services in the industrialized countries are approaching the capacity of copper wires but there is always the potential for additional innovation to increase technical barriers. At the same time. traffic. Improved network quality A key aspect of networking to monitor is the quality of the last-mile connection. proliferate and users are able to take advantage of them for ubiquitous connectivity the expectations surrounding a user experience become more demanding. 80 . The individual sources of information (maps. The challenge going forward will be to aggregate them and integrate any ancillary information sources that are required to optimize the user experience. but also facilitate distributed computation on a scale far beyond what is possible today.

The application also needs to assimilate as much context (e. user history) as possible in order to minimize the required input of the user while still tailoring the output to the user‘s expected requirements. e-mail and SMS). It is further complicated by different client platforms and the lack of awareness each device and application may have of the complete user environment.g. voicemail.The challenge is compounded by the fact that users will not necessarily only own and operate a single device. presence. Citrix GoToMeeting and Cisco WebEx.g. The fact that the network characteristics and machine interfaces may vary considerably makes this objective difficult to achieve. location. which also facilitate presentation and whiteboard sharing. While these technologies may ultimately rely on some peer-to-peer protocol for efficient transfer of high-volume data there will always be a need for central services that provide 81 . There is an increased need for synchronization while respecting authoritative sources of user information. And yet they will expect a consisted experience and unified information view regardless of the location and terminal that they use.g. Unified communications As the Internet gradually replaces most other forms of networking. service providers are seeking to leverage its versatility to integrate both real-time communication services (e. voice and video) and unified messaging services (e. There are also number of collaboration tools. such as Microsoft LiveMeeting.

extensibility. Unfortunately. However. Rather. 82 . efficiency and mobility. the most compelling reason that many organizations will adopt IPv6 is not because of its benefits or any specific limitation that they will encounter with IPv4. the address space is nearing exhaustion – in some countries more than in others.directory information and are able to broker or proxy connections where direct network connectivity is not possible. 2009). There has been a substantial amount of discussion about IPv6 in the IT industry over the past ten to fifteen years with many predicting its imminent arrival earlier than was realistic. Google offers search services over IPv6 but doesn‘t support other products over the protocol. IPv6 also offers numerous benefits in terms of automated management. Beyond merely lifting size limitations. this may have encouraged the illusion that it will never be necessary or viable. Nonetheless. Since it is enabled by default on most common operating platforms it is irresponsible for IT security to ignore it and any threats that may be channeled through it. It is true that network address translation dramatically increased the potential size of an IPv4 Internet. or indeed the protocol of choice. IPv6 3Tera has already demonstrated database replication between applications running in two different datacenters using an IPv6 VPN tunnel created with 3Tera's new VPN catalog appliance(Romanski. even so. So. the need to address IPv6 has become urgent in recent years simply because it is already active on most networks. there is still some way to go before IPv6 becomes ubiquitous. while there is some activity among the cloud service providers.

dynamic renumbering and stateless auto-configuration may simplify an automatic deployment. global computing and public computing). For instance. Folding@home. 83 . It is based on peerto-peer protocols for voice and video over IP. Peer networks are similar to cloud computing in that their attractiveness is highly correlated to the last-mile bandwidth. SETI@home. PEER-TO-PEER NETWORKING Peer-to-peer networking isn‘t new.Regardless of the reason why an enterprise chooses to evaluate IPv6.There has already been some interest in volunteer computing (also called peer-to-peer computing. While they are currently an effective complement to data-center-based services it is quite possible that they will eventually emerge as a competitor to centralized services. The first networks involved peer exchanges. peer networks may benefit from the growing power of home PCs. it is worth considering that it also includes several benefits that can be relevant to cloud computing. Mobile IPv6 facilitates transfer and load balancing of virtual machines. The terms became popular with file and media sharing networks such as KaZaA and later BitTorrent. Einstein@home or AQUA@home leverage the collective power of consumer infrastructure. For example. Skype has become a household term even among the technically illiterate. Unlike cloud computing. They are also network-intensive and therefore only very effective on broadband connections. And the increased address space enables direct end-to-end connectivity between end-points in different clouds without any special requirements from the service providers.

Peer networks can accomplish the same purpose as traditional content delivery networks. If a proper incentive system is devised the potential is significant. But their potential scale with the correct 84 . In small volume they are not sufficiently reliable.However. the business model is still evolving that would enable any service level beyond best effort.