Which of the following cloud computing characteristics gives customers elasticity

9.3.3.1.1 On-demand self-service

On-demand self-service allows customers to use cloud computing as required without human contact between consumers and service providers. Using the features of on-demand self-service, consumers can arrange various cloud resources as needed. In addition to being safe and attentive to the client, the self-service system must be user-friendly in order to access the various cloud resources and to track the service offerings effectively. The primary benefit of on-demand self-service generating efficiencies for both consumers and providers of cloud services (Mell & Grance, 2011).

On-Demand Self-Service

On-demand self-service means that a consumer can request and receive access to a service offering, without an administrator or some sort of support staff having to fulfill the request manually. The request processes and fulfillment processes are all automated. This offers advantages for both the provider and the consumer of the service.

Implementing user self-service allows customers to quickly procure and access the services they want. This is a very attractive feature of the cloud. It makes getting the resources you need very quick and easy. With traditional environments, requests often took days or weeks to be fulfilled, causing delays in projects and initiatives. You don’t have to worry about that in cloud environments.

User self-service also reduces the administrative burden on the provider. Administrators are freed from the day-to-day activities around creating users and managing user requests. This allows an organization’s IT staff to focus on other, hopefully more strategic, activities.

Self-service implementations can be difficult to build, but for cloud providers they are definitely worth the time and money. User self-service is generally implemented via a user portal. There are several out-of-the-box user portals that can be used to provide the required functionality, but in some instances a custom portal will be needed. On the front end, users will be presented with a template interface that allows them to enter the appropriate information. On the back end, the portal will interface with management application programming interfaces (APIs) published by the applications and services. It can present quite a challenge if the backend systems do not have APIs or other methods that allow for easy automation.

When implementing user self-service, you need to be aware of potential compliance and regulatory issues. Often, compliance programs like Sarbanes-Oxley (SOX) require controls be in place to prevent a single user from being able to use certain services or perform certain actions without approval. As a result, some processes cannot be completely automated. It’s important that you understand which process can or cannot be automated in implementing self-service in your environment.

Cloud Computing and Networking Defined

“The cloud” is such a popular image that it might seem to have lost all precision of meaning. And that may be true in some contexts, but there is actually a very precise definition of just what cloud computing services are. According to the National Institute of Standards and Technology (NIST), cloud computing has five essential characteristics.

These are:

On-demand self-service. A customer can unilaterally (i.e., by themselves without having to ask permission) provision computing capabilities such as server time and network storage. These can be done automatically and without requiring human interaction with a service provider (or an IT department, for enterprise networks).

Wide range of network access. Almost everything, from the databases to the applications that create and process them, are available over the network. These resources are accessed through standard mechanisms and protocols available on various “thin” (or skinny) or “thick” (or full-featured) client platforms, such as mobile phones, tablets, laptops, and workstations. In this context, a Web browser is a form of “universal client.”

Resource pooling. The service provider's or enterprise’s computing resources are pooled to serve multiple users through a multi-tenant model (i.e., many users can access the same location’s resources). These different physical and virtual resources are assigned dynamically according to demand.

Rapid elasticity. The capabilities available to users can be provisioned elastically and released when no longer needed, in some cases automatically. This allows rapid scaling, up or down, according to current demand. To the customer, these capabilities available for provisioning can appear to be unlimited and can be granted in any quantity at any time.

Measured service. Cloud computing systems control and optimize resource use by an abstract metering capability appropriate to the type of service used, such as storage, processing power, network bandwidth, or number of and active user accounts. Resource usage can be monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service.

It’s not that cloud service providers such as Amazon or Microsoft or Google or many others provide these defined capabilities out of the goodness of their hearts. It’s that the rise of modern networking and data gathering an interpretation, especially in the post-Web explosion of plentiful DSL bandwidth to residential users, demands these capabilities if the network service is to survive and grow.

Consider the days of the analog voice network, which were not so long ago. The network bandwidth and switching capability was allocated in fixed units if voice channels (4000 Hz) that could be changed only with a “truck roll” or other form of construction service. The network was designed for peak loads, which in the United States was the five busy days of the voice system (Mother’s Day, Father’s Day, Christmas, New Year’s, and the day after Thanksgiving). For the vast majority of the time, the equipment sat idle. Getting a new telephone installed often took weeks (disconnecting one was easier) and a recently as the 1960s, telephone bills were mailed to subscribers after being typed by a clerk. Data services were provided by essentially applying the modulator/demodulator (modem) technology of a fax machine to the new computer (mainframe computer, not PC) technologies appearing on the scene.

As computers grew in power and networks grew in speed, it became feasible to shuttle data around a network so that applications could analyze it and produce reports on trends that would help plan the business. In fact, the rise of the minicomputer and PC made it possible to distribute the processing power accessible on the network. A technique known as remote procedure call (RPC) allowed applications to run on one computer and call a subroutine function on another, and even access data that came from a database on a third. A whole body of knowledge about distributed computing made it possible to spread tasks across many devices and detect and compensate for duplicate results, missing results, and mis-ordered intermediate results (you can’t compute the average until all the values are summed).

Today, whether it is through Web sales or through statistics gathered from new applications and computers and other end-user devices. How did they get permission to grab all this data? Permissions are often buried deep within an end-user license agreement (EULA), acceptance of which is usually required to use the system or install the app.

In the old days, data sat in centralized server computer databases and was sent to clients for processing. Even an enormous data base like the entire customer data set of a national insurance company, or the set of USA’s Internal Revenue Service (IRS) taxpayers, or the entire data from the US Geological Survey, all could be held inside one room in one building.

Today, databases are still gathered into centralized facilities called data centers. However, the amount of data needed to provide valid results, even over the fastest networks available, could never be sent anywhere except the place where it dwells.

So, in a very real sense, cloud computing is “the application of well-understood distributed computing concepts to amounts of data so large they can never be sent anywhere.” So instead of sending the data to the applications, we have to break up the applications and send them to the data. (We’ve also just taken an informal look at how SDN and NFV fit into the whole cloud computing picture, but we’ll be more systematic about that later in the chapter.)

One of the reasons that the cloud has appeared is that businesses which exploited new digital processing technologies finally caught on to the needs of business cycles that many businesses had known for years. For example, when chip makers first began to manufacture digital watches instead of mechanical analog watches with sweep hands, they churned out the same number month after month and wondered why they had large unsold inventories. Only then did they realize, as the analog watch makers had long ago, that there were only two times that people bought watches in the United States: at Christmas and around Graduation. Once they adjusted production cycles, the wind-up watch quickly became a thing of the past. So in a real sense, cloud computing is the adjustment of computing resources to better reflect the busy and idle times of a business cycle, cycles which can vary greatly from one line of business to another (think of Halloween costumes versus Christmas decorations).

But before we look at SDN and NFV, let’s explore some of the terms that come up with discussions of “the cloud” all the time. That is, the idea of “anything as a service” (XaaS).

Abstract

According to NIST definition of cloud computing, it has five characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service, while mobile computing focuses on device mobility and context awareness considering networking and mobile resource/data access. Mobile cloud computing is usually regarded as building on cloud computing and mobile computing; however, it has some unique features such as service offloading, migration, composition, etc. Mobile cloud computing enriches mobile computing technologies and leverages unified elastic resources of varied clouds and network technologies. This chapter provides an overview of various important concepts that are highly related to mobile cloud computing and illustrate their relations through real-world examples.

2.1 Perspective 1: essential characteristics

Essentially, cloud computing offers the following characteristics as defined by NIST (NIST, 2011):

On-demand self-service: Consumers are able to help themselves and decide which services to subscribe to, and how much to invest—all at the swipe of a credit card or using an online payment system. An IT department can now quickly purchase more resources on-demand to cater to sudden spikes in user load.

Ubiquitous network access: Cloud services hinge on the Internet’s infrastructure, and as such provide a ubiquitous availability of services as long as there is an Internet connection. An USA-based executive can perform his roles during business travel, accessing his company’s online resources hosted in Ireland via the Internet connection in Singapore.

Resource pooling: The combined computational power of large amounts of physical and virtual servers provides a cost-effective pooling of resources. Multitenancy solutions have enabled several organizations to share the same cloud computing resources without worrying about data spilling into each other’s logical boundaries.

Rapid elasticity: Cloud services leverage on technologies such as server and storage virtualization to rapidly meet the rise and fall of user load and service demand. A newly launched business expecting 10,000 customers will be able to handle an unexpected load of 1 million customers without worrying about the need to purchase or set up new servers in short notice. Elasticity also improves the utilization of the cloud resources.

Measured service with pay-per-use: Given the above characteristics, it works for both service providers and consumers to have an easy-to-measure payment scheme mimicking the power utilities and cable television model—pay-per-use. At the appropriate price point, pay-per-use has the potential to alleviate the need for forecasting and planning of resources, and reduce wastage of overheads.

The eagle-eyed reader will observe that the above five points do not point to a new technology paradigm, but an Internet-empowered, high-utilization business concept that simply works.

It replaces the awkwardness of predecessor technologies, such as utility computing and grid computing, as it comes with an easy-to-implement and easy-to-understand business and revenue model. Cloud also reduces the expectations on businesses to forecast demand correctly—which like weather forecasting, is rarely achieved successfully.

Currently, most stakeholders reference the NIST Definition of Cloud Computing (NIST, 2011). Recently, ISO/IEC 17788 (Information technology—Cloud computing—Overview and vocabulary) started defining the cloud computing definitions, but the uptake of the fresh set of definitions remains to be seen.

2.1 Cloud Computing Characteristics

The following are the five main characteristics of Cloud Computing that most people agree upon:

On-demand self-service: Cloud services are on-demand; that is, service consumers can automatically request the service based on their needs, without human interaction with the service provider.

Easy to access standardized mechanisms: NIST refers to this characteristic as broad network access; however, the term “global reach capability” is also used. The idea is that it should be possible to access cloud services through the network using standardized interfaces and access mechanisms. Having global reach capability does not mean that these services must always be accessible from the Internet, because this depends on the deployment model used. However, it should be possible to reach the service globally, when policies allow this.

Resource pooling and multi-tenancy: In Cloud Computing, resources [i.e., storages, processors, memory, network bandwidth, and virtual machines (VMs)] are shared between multiple tenants, and assigned exclusively at run time to one consumer at a time. Assigning resources is done dynamically based on the consumers’ needs. Sharing resources can help increase utilization, and hence significantly reduce the operation cost. Scheduling algorithms can be used to dynamically assign resources to different tenants based on the type of workload, fairness, locality, and many other factors [7,8].

Rapid elasticity: Elasticity is the ability to scale in and out by provisioning resources and releasing them, respectively. Cloud Computing should provide mechanisms to allow quick and automatic elasticity. The large pool of resources in cloud datacenters gives the illusion of infinite resources to the consumers, and elasticity provides the flexibility to provision these recourses on-demand.

Measured service: Providing cloud metrology or mechanisms to measure service usage as well as to monitor the health of services is crucial in Cloud Computing. Measuring services enables optimizing resources and provides transparency for both consumers and providers, allowing them to better utilize the service. Measured services can help in building closed-loop cloud systems that are fully automated.

Auditability and certifiability: Regulatory compliance requires enforcing rules and regulations. Services should provide logs and trails that allow the traceability of policies, so as to ensure that they are correctly enforced.

The list above (except point f) is based on the NIST definition. The list describes Cloud Computing based on what is currently available in the market, and represents the main characteristics of Cloud Computing in general. Many other characteristics can be added to this list in the future. For example, we added “auditability and certifiability” to the list above based on the current regulatory compliance requirements. On the other hand, comprehensive lists of characteristics can be made for each layer and each type of service provided in the cloud environment. For example, at the application level, a possible cloud service characteristic is that a service must be portable, pre-configured, or adaptable.

15.2.2 NIST essential characteristics

The essential characteristics of cloud computing, which make cloud computing different from other computing service models, are as follows [4]:

On-demand self-service: The customers can assign themselves computing resources as needed.

Broad network access: The network can be accessed by mobile phones, laptops, and personal digital assistants (PDAs) as well as more traditional wired connections.

Resource pooling: Computing resources are combined to serve multiple customers with different physical and virtual resources dynamically assigned and reassigned according to demand. For example, the customer has no knowledge of the exact location of their storage, and the location may change.

Rapid and elastic resource provisioning: The quick responses to requests for resources make it appear that they can be purchased continuously and with unlimited quantity.

Metered service at various quality levels: Resource use can be monitored, controlled, and reported, and is billed in proportion to usage, at some level of abstraction appropriate to the type of service.

The essential characteristics that have the greatest effect on data center network design are on-demand self-service, resource pooling, and rapid and elastic resource provisioning. On-demand self-service and rapid and elastic resource provisioning require a flexible, scalable infrastructure, which does not exist in traditional data centers. Resource pooling requires a different design for the resources.

5.2.4 Properties of cloud computing

To be truly attractive, cloud computing needs to satisfy five essential characteristics, set out by the NIST [MEL 11]:

On-demand self-service: the client of a cloud supplier should be able to obtain an extension of resources: processing capacity, storage capacity, network infrastructure, a virtual machine, etc., on demand, and without the need for human interaction with the service provider.

Resource pooling: a given cloud service should be able to serve multiple users simultaneously, using a multitenant model, with physical and virtual resources dynamically assigned and reassigned according to customer demand. The customer has no idea of the location of the requested resource, but may be able to specify a rough location (such as a country or state) in their requests.

Rapid elasticity: the capabilities of the cloud should appear unlimited to the user. These capabilities should be elastically scalable both outward and inward in accordance with demand, whatever the quantity of resources required, and at any time.

Broad network access: capabilities should be accessible over the network through standard mechanisms for any type of thin or thick client terminal (desktop or laptop computer, smartphone, tablet, etc.).

Measured services: resource usage is metered independently for each cloud user, which notably allows pay-as-you-use billing.

What is the Cloud?

The term “cloud” has taken on a life of its own over the last several years. When talking about the cloud there are many different approaches that can be used.

When referring to the cloud the following are the essential characteristics that are needed for something to be referred to as a cloud service:

On-demand self-service.

Broad network access.

Resource pooling.

Rapid elasticity.

Measured service.

Some of the service models that are used to deploy the cloud are:

Software as a Service (SaaS).

Platform as a Service (PaaS).

Infrastructure as a Service (IaaS).

Within those service models the following are the deployment models that can be used:

Public/vendor cloud.

Private cloud and virtual private cloud.

On and off-premises.

Hybrid cloud.

Community cloud.

While cost is definitely one valuable reason for moving to the cloud the two big drivers typically focus around IT Efficiency and Business Agility:

IT Efficiency:

Enables a variable cost model for IT.

Minimizes overall IT costs—Shift CAPEX to OPEX.

Improves infrastructure resource deployment and utilization through virtualization.

Provides a flexible, reusable application development model.

Business Agility and Market Competitiveness:

Enables quicker “time-to-market.”

Rapid application deployment.

Reduced infrastructure setup/configuration.

Support for large-scale parallel programming.

Reduces switching costs associated with changing business strategies.

Alternatives for cost reduction efforts—allows for outsourcing segments of IT.

While this book is not about general cloud implementation, the focus is on the impact it has on security and overall APT.

One of the biggest reasons on why the cloud represents an exposure to many organizations is proper planning is not performed. If organizations actually understood the pros and cons, analyzed the exposures, and properly planned, the cloud can be an effective, cost-effective secure method for providing services to an organization.

Cloud computing characteristics

The main objective of consolidating and virtualizing the various cloud computing building blocks in Figure 1.32 is to attain an SOI with the following characteristics:

On-demand self-service: An end-user can unilaterally provision computing capabilities, such as server settings and network storage when needed, without any interaction from the provider's IT administrator.

Universal network access: Capabilities are available over the network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms, such as mobile phones, laptops, netbooks, tablet computers, personal digital assistants (PDAs), and so on.

Resource pooling: The provider's computing resources are pooled to serve multiple end-users using a multitenant model, with different physical and virtual resources dynamically assigned and reassigned according to the end-user needs. Examples of such resources include storage, processing, memory, network bandwidth, and virtual machines. There is a degree of location freedom (or independence) in that the end-user generally has no notion of the exact location of the provided resources but will be able to access these resources from an intranet if the end-user is an internal staff member or access from the extranet/Internet if the end-user is an external party.

Rapid elasticity: Capabilities can be rapidly and elastically provisioned (in some cases automatically) to quickly scale out and rapidly released to quickly scale in. To the end-user, the capabilities available for provisioning often appear to be unlimited (or boundless) and acquirable.

With these essential characteristics defined, it is time to delve into the various cloud computing deployment and service models.