CHAPTER 1
Basics

Upon completion of this chapter the reader will understand:

1. The standard definition, characteristics, and benefits of cloud computing
2. The key roles in the cloud computing ecosystem
3. Key concepts of application, demand, supply, quality, and fungibility

This section reviews the following key concepts that are used extensively in this work:

• Cloud computing fundamentals (Section 1.1)
• Roles in cloud computing (Section 1.2)
• Applications (Section 1.3)
• Demand, supply, capacity, and fungibility (Section 1.4)
• Demand variability (Section 1.5)

1.1 Cloud Computing Fundamentals

Cloud computing is defined by ISO/IEC 177881 as a "paradigm for enabling network access to a scalable and elastic pool of shareable physical or virtual resources with self-service provisioning and administration on-demand." ISO/IEC 17788 specifies the following six key characteristics of cloud computing2:

1. Broad network access - "A feature where the physical and virtual resources are available over a network and accessed through standard mechanisms that promote use by heterogeneous client platforms. The focus of this key characteristic is that cloud computing offers an increased level of convenience in that users can access physical and virtual resources from wherever they need to work, as long as it is network accessible, using a wide variety of clients including devices such as mobile phones, tablets, laptops, and workstations" (ISO/IEC 17788). Operationally, this means that end users access cloud-based application services via commonly available wireless and wireline IP networks.
2. Measured service - "A feature where the metered delivery of cloud services is such that usage can be monitored, controlled, reported, and billed. The focus of this key characteristic is that the customer may only pay for the resources that they use. From the customers' perspective, cloud computing offers the users value by enabling a switch from a low efficiency and asset utilization business model to a high efficiency one" (ISO/IEC 17788). When cloud customers pay only for resources that are used, application services that are engineered so cloud resource usage tracks with application service usage which tracks with application revenue can reduce business risk by better linking application service provider's costs with application service revenues.
3. Multi-tenancy - "A feature where physical or virtual resources are allocated in such a way that multiple tenants and their computations and data are isolated from and inaccessible to one another" (ISO/IEC 17788). Multi-tenancy enables infrastructure service providers to maximize resource sharing and boost their capacity utilization.
4. On-demand self-service - "A feature where a cloud service customer can provision computing capabilities, as needed, automatically or with minimal interaction with the cloud service provider. The focus of this key characteristic is that cloud computing offers users a relative reduction in costs, time, and effort needed to take an action, since it grants the user the ability to do what they need, when they need it, without requiring additional human user interactions or overhead" (ISO/IEC 17788). This means that application service providers and/or automated systems working on behalf of those application operators can install, configure, and provision cloud resources to serve their applications in real time. On-demand self-service of capacity planning and fulfillment actions, coupled with rapid elasticity, enables significant reductions in fulfillment times for capacity change actions compared to traditional deployments.
5. Rapid elasticity and scalability - "A feature where physical or virtual resources can be rapidly and elastically adjusted, in some cases automatically, to quickly increase or decrease resources. For the cloud service customer, the physical or virtual resources available for provisioning often appear to be unlimited and can be purchased in any quantity at any time automatically, subject to constraints of service agreements. Therefore, the focus of this key characteristic is that cloud computing means that the customers no longer need to worry about limited resources and might not need to worry about capacity planning" (ISO/IEC 17788). Application service providers and/or automated systems working on their behalf can allocate and release infrastructure resources on-the-fly, thereby enabling applications to transform from allocating and configuring capacity based on peak forecast demand (which may never even be approached) to just-in-time, demand-driven capacity configuration.
6. Resource pooling - "A feature where a cloud service provider's physical or virtual resources can be aggregated in order to serve one or more cloud service customers. From the customer's perspective, all they know is that the service works, while they generally have no control or knowledge over how the resources are being provided or where the resources are located. This offloads some of the customer's original workload, such as maintenance requirements, to the provider" (ISO/IEC 17788). Resource pooling, coupled with multi-tenancy, enables cloud service providers (CSPs) to leverage economies of scale to boost operational efficiencies beyond what has traditionally been feasible.

Figure 1.1 offers a simplified view of cloud computing: many different applications simultaneously share a large and flexibly configured pool of compute, memory, storage, networking, and functional component resources offered by a CSP. Coupling on-demand self-service and rapid elasticity enables application service providers to fundamentally transform their businesses from configuring application capacity to meet peak forecast demand to just-in-time, demand-driven capacity management. Using measured service as the basis of charging application service providers for their resource usage enables at least a portion of the application provider's costs to shift from being fixed and independent of usage to variable so that resource charges are tied to resource usage, which is tied to application service usage, which should be tied to application revenue.

Figure 1.1 Cloud Computing in a Nutshell

Figure 1.2 visualizes key high-level business benefits of cloud computing:

1. Lower infrastructure CapEx - the peak aggregate demand for compute, storage, and networking resources is smaller than the sum of peak demand of each individual application. Infrastructure resource sharing enables less total hardware to be deployed than with traditional operation models, and thus capital expense (CapEx) can be saved. Also, consolidating operation of that shared infrastructure enables efficiencies that can reduce operating expense (OpEx).
2. Infrastructure available on demand - the ability of application service providers to allocate and release infrastructure resources on-the-fly enables agile, just-in-time application capacity management to minimize the twin business risks of over investing in application capacity that is never used (if demand forecasts are too optimistic) or being unable to serve customer demand (if demand forecasts are too conservative).
3. Lower infrastructure OpEx because a single infrastructure service provider can operate and maintain shared infrastructure cheaper than each application service provider can operate and maintain their own private infrastructure. In addition, consolidating active resource loads onto the smallest number of physical elements enables unneeded capacity to be turned off.
4. Application OpEx for infrastructure tracks with resource usage thereby derisking application service provider businesses by having their OpEXs for resource capacity track with service usage, and thus hopefully with revenue.

Figure 1.2 High-Level Benefits of Cloud Computing

Cloud enables disruption of IT service management in several ways:

• Encourage agile, incremental development and deployment models - these practices can reduce the organization's business risks around development activities by enabling applications and services to be developed and deployed via smaller project tasks which can generally be better managed rather than relying on massive software projects that are harder to manage.
• Elastic capacity enables new operational models for application release management - infrastructure available on-demand means that instead of relying on complex software upgrade, update, or retrofit procedures that must execute "in-place" on traditional hardware, an independent set of infrastructure resources can be allocated to host the new software upgrade, update, or retrofit and that release maintenance action can largely proceed without impacting the production instance thereby both reducing service risk for users and operational complexity (and presumably costs) for application service providers.
• Fungibility of virtualized...