Daryl C. Plummer, Thomas J. Bittman, Tom Austin, David W. Cearley, David Mitchell Smith of Gartner give an overview of the emerging landscape that is cloud computing

The cloud is emerging as the latest way to approach alternative delivery models for IT capabilities. It is a way of delivering IT-enabled services in the form of software, infrastructure and more. This research examines the definition of cloud computing and how it will evolve.

Cloud computing is an alternative delivery and acquisition model for IT-related services. This paradigm will shift the way purchasers of IT products and services contract with vendors and the way those vendors deliver their wares. However, there are also risks and barriers to the emergence of this model. A definition of cloud computing is a style of computing where massively scalable IT-enabled capabilities are delivered 'as a service' to external customers using Internet technologies. This leads to the industrialization of IT and will alter the way many organizations deliver business services that are enabled by IT. It also leads to new ways of acquiring and using technology that not all businesses will be ready for.

Throughout the history of business, delivery of shared business services has been a key enabler of growth and a way to more consistently penetrate larger and wider customer bases. For example, shared product delivery services have reduced costs for shippers and consumers, and shared customer service processes have shaped an entire era of responsive call centers.
If we go a bit further back, the advent of industrialization offered the world the ability to deliver a wide range of products at a price that was low enough to make available, to average people, products that were previously only available to the wealthy or to governments and businesses. It was only a matter of time before shared IT services gained significant penetration to foster dramatic shifts in the IT industry. The use of virtualization technologies, open source, service-oriented architectures and widely available computing standards, combined with the pervasiveness of the global Internet, is making computing-related services generally available to the world at reduced costs and massive scale. Because of this, new IT service delivery and acquisition models are emerging. One such model has been termed ‘cloud computing.’ In this research, we examine Gartner's definition of cloud computing and its place in the changing landscape of IT.

What Is Cloud Computing?
Gartner defines cloud computing as "a style of computing where massively scalable IT-enabled capabilities are delivered 'as a service' to external customers using Internet technologies."

If we break down Gartner's definition, what we find is a set of mutually supportive concepts. First and foremost is the concept of delivering services (that is, results as opposed to components).

Implementation doesn't matter as long as the results of the implementation can be defined and measured in terms of a service with associated service-level requirements. Included in this concept is payment based on usage, not on physical assets. The payment can be subsidized (for example, by advertising) or paid directly by the customer. The second concept is that of massive scalability. Economies of scale reduce the cost of the service. Implicit in the idea of scalability is flexibility and low barriers to entry for customers. Third, delivery using Internet technologies implies that specific standards that are pervasive, accessible and visible in a global sense are used. Finally, these services are provided to multiple external customers, leveraging shared resources to increase the economies of scale.

Scalability vs. Elasticity
The question of scalability is often discussed in relation to the cloud. Gartner's definition mentions massive scalability but does not explicitly mention the concept of elasticity. Scale is an aspect of performance and the ability to support customer needs. The concept of elasticity is related to the ability to support those needs in large or small scale at will. The key issue with elasticity is the ability for a system to scale both in an upward direction (for example, to millions of users) and in a downward direction (for example, to one user) without disrupting the economics of the business model associated with the cloud service. Typical enterprise-class systems are scaled in the upward direction, but the cost of running that scaled complex for one or two users would be prohibitive given the cost of operating and maintaining the environment. However, global-class providers like Google, eBay or Zoho have a model that is not primarily based on the cost of the infrastructure and operations or software licenses and maintenance, but leverages ad revenue and other mechanisms to support their existence. This means that the number of users and what they are doing, while extremely relevant, can be detached from the operating economics of the cloud provider.

In addition to the issue of elasticity is the question of whether something that does not provide massive scalability is to be considered cloud or not. The definition is not intended to be a threshold that determines inclusion in the cloud model. Instead, it is intended as a guideline for the relative ‘cloudiness’ of a particular solution. This means that massive scale is not an ultimate attribute for cloud-computing providers — It is an indicator of where they are in the relative breadth of their cloud solutions. As cloud computing evolves, a small number of massive IT providers will emerge supporting massive workloads, massive data manipulation and general-purpose services. Their differentiation will primarily be economies of scale. However, there will also be a ‘long tail’ of midsize and even relatively small providers differentiating on leading-edge and special-purpose technologies, thriving on continued IT innovation and maintaining price pressure on even the most massive providers.

Changing How Services Are Delivered
During the past 15 years, a continuing trend toward industrialization of IT has grown in popularity.

IT services delivered through hardware, software and people are becoming repeatable and usable by a wide range of customers and service providers. This is, in part, because of commoditization and standardization of technologies, virtualization and the rise of service-oriented software architectures, and, most importantly, the dramatic growth in popularity/use of the Internet and the Web. These things, taken together, constitute the basis of a discontinuity that amounts to a new opportunity to shape the relationship between those who use IT services and those who sell them.

What the discontinuity implies is that the ability to deliver specialized services in IT can now be paired with the ability to deliver those services in an industrialized and pervasive way. The reality of that implication is that users of IT-related services can focus on what the services provide to them rather than how the services are implemented or hosted. Just as utility companies sell power to subscribers and telephone companies sell voice and data services, IT services such as network security management, data center hosting or even departmental billing can now be easily delivered as a contractual service. The buying decision then shifts from buying products that enable the delivery of some function (like billing) toward contracting with someone else to deliver those functions. Certainly, this is not new, but it does represent a different model from the license- based, on-premises models that have been dominant in the IT industry for so long. Names for this type of operation have come into vogue at different times. Utility computing, software as a service (SaaS) and application service providers (ASPs) all have their places in the pantheon of industrialized delivery models. However, none has garnered widespread acceptance as the central theme for how any and all IT-related services can be delivered to the world.

The types of IT services that can be provided range far and wide. Compute facilities (for example, Amazon's Elastic Compute Cloud [EC2]) provide computational services so that users can use CPU cycles without the need to buy computers. Storage services (for example, Amazon's Simple Storage Service [S3]) provide a way to archive data and documents without the need to continually grow farms of storage networks and servers. SaaS companies like salesforce.com offer CRM services through their multitenant shared facilities to clients to manage their customers without ever buying any software. This represents only the beginning of options for delivering complex capabilities of all kinds to businesses and individuals alike. Even client computing can be provided as a service (through hosted desktop and virtual-machine technologies), thus potentially removing the need for a PC altogether. These things, and more, are increasingly falling under the banner of cloud computing.

Cloud and Related Concepts
There are many concepts that seem similar to cloud computing but are, in fact, complementary to it. If we examine a few of them, we can establish how these concepts build on one another rather than just being alternative names for the same phenomenon.

SaaS: With cloud computing gaining substantial buzz in the IT industry, some vendors have transformed their positioning from SaaS providers to cloud-computing providers without changing one element of their offerings. When comparing the two definitions, it becomes clear that cloud computing is a necessary underpinning for a provider to deliver a scalable SaaS offering to the market.

Massively scalable IT-related functions are a concept that goes directly to a provider's ability to deliver a single set of common code and data definitions, which are consumed by all contracted customers. This is fundamental for any SaaS application.
Web Platforms/Cloud Platforms: A Web platform provides programmatic access to Web-based capabilities that act as a foundation to create a composite application and/or business process.

Capabilities (that is, services) may include infrastructure, applications, content, application components, business processes or ecosystem management. The programming model is built on Web-oriented architecture and principles. This is complementary to cloud computing in that the Web platform (also called a cloud platform) is used to enable the provisioning of services through Web-based architecture to the cloud. There is not one Web platform but multiple Web platforms.
Utility: Wikipedia defines utility computing as: "The packaging of computing resources, such as computation and storage, as a metered service similar to a physical public utility (such as electricity, water, natural gas, or telephone network). This system has the advantage of a low or no initial cost to acquire hardware; instead, computational resources are essentially rented. Customers with very large computations or a sudden peak in demand can also avoid the delays that would result from physically acquiring and assembling a large number of computers."

Grid: A grid is a collection of resources (owned by multiple organizations) that is coordinated to enable the resources to solve a common problem. They may be divisions of one company or distinct legal entities. Three conceptual forms of grids are computing grids (the most common today), data grids and collaboration grids.

Real-Time Infrastructure (RTI): An RTI is an IT infrastructure shared across customers, business units or applications where business policies and SLAs drive its dynamic and automatic optimization to reduce costs while increasing agility and quality of service. RTI includes the concepts of service orientation and shared resources for multiple customers. Add access through Internet technologies and an RTI is a cloud-computing service for infrastructure. Essentially, RTI, in one form or another, will be the ‘engine room’ for cloud-computing providers.

Conclusion
Cloud computing is an evolving concept and will take many years to fully mature. Our definition establishes the Internet as part of that evolution and allows for the proliferation of IT services in a globally pervasive way. The relationship between buyer and seller is thus redefined, and the way in which individuals or IT departments gain value from technology is altered. Customers seeking to shift costs from on-premises implementations to delivery of services from providers who offer SLAs and performance guarantees will benefit from a cautious approach to a cloud model.

Vendors selling IT technologies and services will begin to shift to cloud platforms and cloud- based services as primary marketing messages in the next few years or run the risk of being marginalized as their products sell less to users who are buying more from cloud-computing service providers. The cloud-computing model is not just the next generation of the Internet. When organizations cross the threshold between "the Internet as a communications channel" and "the deliberate delivery of service over the Internet," everything begins to change. The Internet alone does not do that.