Handbook of Cloud Computing: Basic to Advance research on the concepts and design of Cloud Computing

By Dr. Anand Nayyar

The book “Handbook of Cloud Computing” provides the latest and in-depth information of this relatively new and another platform for scientific computing which has great possibilities and high future prospects to become ten folds in near future. The book covers in comprehensive manner all aspects and terminologies associated with cloud computing like SaaS, PaaS and IaaS and also elaborates almost every cloud computing service model.

The book highlights several other aspects of cloud computing like Security, Resource allocation, Simulation Platforms and futuristic trend i.e. Mobile cloud computing. The book will benefit all the readers with all in-depth technical information which is required to understand current and futuristic concepts of cloud computing. No prior knowledge of cloud computing or any of its related technology is required in reading this book.

• Language: English
• Publisher: BPB Online LLP
• Release date: Jul 22, 2019
• ISBN: 9789388511506

Book preview

CHAPTER 1

Introduction to Cloud Computing

Learning Objectives

In this chapter, the concept of cloud computing is introduced and significantly highlighted, regarding its role towards bringing a significant shift in the world of technology.

The chapter dives into various types of layer’s and deployment models present in cloud computing.

The chapter lists and explains the benefits and limitations associated regarding shifting to cloud technology.

The chapter sheds light on the various cloud computing service providers in present world.

1.1 Evolution and History of Cloud Computing

Before the rise of cloud computing, there existed a centralized storage containing all the software applications, data, and services as a part of Client/Server computing on the server side. Hence, for a single person to access data, he/she needs to gain access to the server. Then after, the concept of distributed computing was introduced, resource sharing was made possible. This later on led to the evolution of cloud computing. Cloud Computing came into existence in 1950s when mainframe computers [1] were accessed via dummy terminals into a central computer. It was done by users to gain access. The need for the idea of sharing resources was felt as the excessive costs of mainframes were not economically feasible. Hence, there was an urgent need to reduce costs.

In the 1970s, IBM introduced an operating system called VM and the concurrent operation of more than one OS was made possible. VM’s can allow Guest Operating Systems [5] to run on it, having their own infrastructure and memory and resource sharing also made possible. This made the concept of virtualization quite popular. In the 1990s, the telecom companies began providing virtualized private network connections [2], with better service and quality than point-to-point services at less cost. This made companies to offer shared access to users in a single infrastructure. The evolution of cloud computing can be diverted into three main phases:

The Idea Phase: This started in the early 1960s with the rise of service and grid computing and continued until the pre-internet era.

The Pre-cloud Phase: This stage initiated in the year 1999 and continued till 2006. In this, the internet was used as the machinery to provide applications as a service.

The Cloud Phase: The concept of IaaS, PaaS, and SaaS was clearly identified and became practical in the year 2007. Since then the cloud computing has continuously evolved changing the face of world via resource sharing and transforming end-user computing.

Cloud Computing is not a new idea. The evolution of cloud computing started back in history during the 1900’s when telecommunication companies wanted to transform their communication networks. Since then there were many technological advancements, and this consistent innovation is what we see today as modern-day cloud computing.

The Early 1990s

Traditionally, telecom companies only provided point-to-point connection networks to the clients. In the early 1990s, though, they started increasing their services by providing virtual private networks. This permitted the telecommunication companies to deliver the identical class of service at a small portion of the fee with the capability of better resource consumption improving the proficiency of bandwidth.

The Late 1990s

During the starting years of cloud evolution, it was seen as a characteristic measure for computing space between the service provider and the client. In year 1997, it was demarcated that cloud computing is an innovation and it will define new frontiers not just in technical but commercial foundation’s too. Professor Ramnath Chellappa elucidated this innovation as a revolutionary idea for businesses.

During the next half of the 1990s, businesses started gaining an improved considerate of cloud computing and its helpfulness in giving higher explanations and facilities to customers while severely civilizing interior competences. In the year 1999, Salesforce.com developed one of the chief powerhouses in the cloud domain, found the notion of utilizing the Internet in transporting business-level applications to final clients. The consumers can access the applications via the Internet and companies can help make business on-demand basis.

The Early 2000s

Salesforce.com introduced a novel idea and Amazon.com demonstrated it’s ability to outshine in cloud market by providing web-based services in 2002. The modernization of data center’s was first done by Amazon which was operating only about 10% of their volume at a certain time. Then, Amazon comprehended that the prototype can help them use their current capacity with high productivity [3].

The Late 2000s

In the meantime, Google became an important competitor in the Internet business market. In 2006, Google Drive conveyed the control of cloud computing and text distribution straight to end clients. In 2006, Amazon Web Services launched S3 and EC2, a new market for cloud computing was evolved. In 2009, Rackspace with the development of Swift replaced Mosso Cloud files with OpenStack Object Storage Project. The year 2010 marked significant with cloud developments using Agile and DevOps concept.

The future of cloud computing seems bright. Many organizations have indulged in some significant predictions. The networking giant, Cisco forecasts a factor of 12 increases for data centres traffic between 2010 and 2015 [4]. Forrester assumes the cloud market to rise from $40.7 billion in the year 2011 to above $241 billion forecast for the year 2020. The data analysis experts at Gartner expects a $150 billion increase in business by the year 2014 and for medium enterprises, it will be over $100 billion by the year 2014 as per AMI partners, [26].

1.2 Introduction to Cloud Computing

Cloud computing is regarded as on-request distribution of power, database space, applications, and other resources using a service platform via Internet. Basically, cloud computing is a form of subcontracting of applications where end-users enjoying its benefits without worrying with regard to storage space and power consumption.

National Institute for Standards and Technology (NIST) [8] defined cloud computing as: a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.

Berkeley [7] defined cloud computing as: to include application software delivered as services over the internet, the hardware, and systems software in the data centres that facilitate these services.

Traditional business applications are very complex and costly. The quantity and variation of hardware and software essential to track them are unnerving. There is a need for professionals in installation, testing, securing, and updating. Hence, a lot of efforts are required for suitable handling of cloud infrastructure. The support of cloud computing helps to eliminate headaches with regard to storing one’s own data, managing hardware and software as it becomes the accountability/responsibility of cloud vendor. The shared structure becomes a utility: Paying only for services, automatic up gradation, and scaling of resources.

In cloud services, all the work becomes slightly contrary, provisional on the part of service provider. Many of them provide a responsive, browser-based dashboard that makes it possible for IT professionals and developers to order resources and managing accounts quite easily. Some of the cloud services are planned to work with REST APIs [6] and a command-line interface (CLI), providing developers with several choices. Figure 1.1 highlights the importance of cloud computing. In the recent era, businesses are shifting their operations to the cloud, like customer relationship management (CRM), HR, accounting, etc.

Some of the things that can be performed via cloud are enlisted as follow:

Creating new applications and services

Storage, back-up, and data recovery

Hosting website and blogs

Streaming live video and audio

Software delivery on demand

Data Analysis and predictions

Figure 1.1 Importance of Cloud Computing

Based on the services, cloud computing offers models such as Infrastructure as a Service (IaaS): providing hardware, software, storage, and servers, Software as a Service (SaaS): to access cloud applications and other software, Platform as a Service (PaaS): the platform to code, run, and deploy apps. It also includes deployments models such as public, private, community, and hybrid. Cloud Computing offers many benefits to users in terms of flexibility for program customization’s, multi-tenancy to deliver applications to large number of consumers, reliability for relaxed customer support, accessibility to integrate with IoT applications. The detailed information is highlighted in section 1.3.

1.3 Why Cloud Computing is Becoming Highly Important?

End users use webmail, photo storage sites like Gmail, Google Photos, document storage services like Drop box and Google Docs and social networking sites like Facebook, Twitter. One thing they all have in common is cloud computing services. Cloud Computing is used by large and small businesses and even individuals also. It is all about transporting computing services distantly, via Internet. It’s an expedient collective pool of resources that can be scaled up or down subject to the varying needs of an organization.

Cloud services are important as they can lessen the cost and intricacy of possessing and operating computers and other network resources. Some of the benefits include are- low front-up costs, quick return on investment, easy customization, and fast deployment. This is because cloud users do not have to finance in infrastructure, hardware, and licensing. In addition, cloud suppliers that have an expert in a specific domain can transport cutting-edge services that the company might not be able to develop alone. The cloud is steadfast, that it allows access to programs and documents from anywhere in the world. It is often deliberated effective because it permits organizations to free up resources to emphasize improvement and development.

Another prospective advantage is that the information can be given better protection in the cloud. Cloud computing [3] helps improve efforts in building privacy protection mechanisms via use of efficient security systems. It enables IT procurement and enhancements, which permit adjustments with regard to procedures based on data understanding. Cloud computing is a boost for all small and medium-sized enterprises. Research demonstrates that SMEs can save up to 70% of their IT budgets if they transfer their IT organization to the cloud which also provides backing up data in case of any failure.

The reasons why Cloud Computing has become the need of the hour:

Flexibility: Cloud services are perfect for businesses with developing or shifting bandwidth needs. It is quite easy to scale up the service by increasing the storage capacity or even scale sown when the demands are met. This scope of agility can contribute businesses with a real benefit over contenders.

Adversity retrieval: Big Businesses participate in healthy adversity recovery, but for smaller ones lack the required costs and skill; this is an ultimate truth. Cloud Computing is now serving more organizations buck that drift. It is more ideal for small organizations to prevent investing in large recovery costs.

Automatic software updates: Cloud computing servers are not in reach of the organizations instead the contractors operate it for them including time to time software updates and even security patches. It helps companies to focus on building their business and not taking stress for maintaining their hardware systems.

Investment-expenses free: It comes with a friendly subscription model to pay for the services needed and no need to take care of hardware installations and maintenance. Cloud computing provides the option to ease management operations and rapid project building.

Improved partnership: Cloud Computing [3] makes it easy to work together, easy workflow and file sharing are possible in real time. The contribution of each member is clearly visible to the team and documents can be edited and shared anytime, from anywhere.

Security: It also guarantees a more secure atmosphere, acknowledgments to improved resources for safety and data centralization.

1.4 Cloud Types

Cloud models are a fundamental entity that describes the various components and elements that constitute as a part of the cloud organization. Figure 1.2 shows how service models in cloud work to achieve client and vendor requirements successfully in low time and cost.

Figure 1.2 Cloud Computing Service models

Source: https://www.ibm.com/blogs/cloud-computing/wp-content/uploads/2015/04/Cloud-computing-service-models.png

1.4.1 The NIST Model

The National Institute of Standards and Technology (NIST) [9], an organization under the US Department of Commerce is accountable for descriptive & describing standards in Science and Technology. The modern NIST model of cloud computing necessitates virtualization and employs the notion of multi-tenancy.

Cloud computing models enable suitably, permission to on-request access to a collective pool of customizable resources including storage space, network servers, and applications. It can be quickly deployed with insignificant efforts, interaction with the service provider. The cloud model encourages accessibility and consists of five crucial features:

On-request self-service

Wide-ranging network access

Resource sharing

Rapid elasticity

Measured Service

Figure 1.3 shows the NIST model in three dimensions consisting of service models, deployment models, and essential features.

Figure 1.3 NIST Cloud model

Source: https://www.katescomment.com/wordpress/wpontent/uploads/2010/02/CloudCube1.png

Key facilitating technologies present in the NIST model comprise: (1) faster wide-area networks, (2) great, low-cost server computers, and (3) high-performance virtualization for product hardware.

Some of the important features present in the NIST model are as follows:

On-request self-service: It provides provision of cloud abilities to the customer, such as server time and storage. It is totally based on-demand as per need and requirements without any sort of human intervention with the service provider.

Wide-ranging network access: Competencies are accessible over the network and retrieved via typical tools that promote the use by varied client platforms (e.g., mobile devices, laptops, and Personal Digital Assistants).

Resource sharing: The provider’s resources are shared to help numerous customers using a model as multi-tenant, with diverse physical and simulated resources virtually allocated and reallocated according to customer’s needs. The customer has generally no mechanism or idea of the exact location of the provided resources, this brings about location independence but sometimes there needs to be information given at a higher level of intellection. Examples include space, processing, memory, bandwidth, and VMs.

Fast elasticity: Abilities can be quickly accessed, in some cases routinely, to rapidly scale out and quickly unrestricted to speedily scale in. To the customer, the utilities seem to be limitless and can be obtained in any measure at any given time.

Measured Service: Cloud services are metered and can be purchased by the user as per rate basis at any given time. Monitor service usage can be observed, measured, and testified providing clearness for both the worker and customer of the consumed facilities.

1.4.2 Cloud Cube Model

There is an open-group association Jericho Forum [10] & their emphasis is on how to safeguard and protect cloud network. They put an accelerative model that aids to classify a cloud-based network on four-dimensional aspects. The cloud cube model is highlighted in figure 1.4.

Figure 1.4 Cloud Cube Model

Source: [https://www.w3schools.in/cloud-computing/cloud-cube-model/]

Internal/External: It describes the location of data storage. If the location of data storing is confidential from the side of data owner, then the prototype is considered to be internal. Inversely, the model is said to be external. External Data cloud storage is more secure than internal data storage. The model combines both internal and external parameters to make cloud model overall highly secure.

Proprietary/Open: It describes the possession of cloud utilities. This factor highlights the mark of adaptability, i.e. the degree of transfer of information between the registered system and other cloud modules, the capability of converting information from a module present in the cloud to another deprived of any restraint. Proprietary means that a cloud vendor grasps the proprietorship of services to provide cloud services. The skills used in the public cloud are usually open and unchanging, meaning more accessible vendors and less restriction on data sharing and integration with corporate associates. Unverified but maximum, exposed clouds can stimulate successful combination between manifold enterprises.

Parameterised/De-parameterised: It defines the architectural mindset of security shield, i.e. a consumer application within the security boundary. Parameterised means that client’s application functions inside outdated IT safety border signalled by a firewall that prevents the combination of different safety sectors. In fact, clients operating some applications inside of safety boundary can spread/shrink their application border to/back from outside environment using Virtual Private Network. De-parameterised is the fade way of outdated IT safety border and the contact of a client’s application procedure.

Insourced/Outsourced: The last parameter defines two positions in each of the eight cloud forms. Insourced defines that the facility is accessible by an organization own personnel, and Outsourced means that the facility is offered by a third party. This answers the question, who is going to manage or build cloud services. In this cloud model, other characteristics like Offshore and Onshore parameters are also applicable.

1.4.3 Deployment Models

The different cloud deployment models are:

Private cloud: used completely by a single corporate or organization and can be physically situated on the company’s on-site data centre.

Public cloud: possessed, functioned and hosted on the foundation of the third-party service provider to transport their resources like servers and storage space over the web.

Community cloud: pooled by a set of organizations of similar trades and circumstances with comparable necessities i.e. task, safety, agreement and IT rules.

Hybrid cloud: grouping of more than one model together such as public, private or community cloud to transport information and programs and making use of regular technology without disclosing any sort of cloud information.

More detailed information regarding cloud deployment models can be observed at Section 1.5.

1.4.4 Service Models

The different cloud service models are: Infrastructure as a Service (IaaS) : provides comprehensive elasticity to customers in choosing desktops, VPN’s, disk drives, mailing networks, DNS etc., Platform as a Service (PaaS): providing platform to transport development/operating atmosphere as a facility having physical servers, databases and online servers, and Software as a Service (SaaS): to deliver a single application through the browser to numerous users via vendor who sells an application to consumers on license basis.

Some of the services of various Service Models in different sectors are shown in table 1.1.

The detailed explanation for these service models is present in Section 1.5.

Table 1.1 Some of the services in different sectors include:

1.5 Layers and Types of Cloud

In the previous section, we threw light at the different levels of abstraction present in the cloud. They were IaaS, PaaS, and SaaS. We also highlighted different kinds of deployment models present in the cloud, which can be the private cloud, public cloud, community or the hybrid cloud.

In this section, we take a deeper look towards different layers and types of cloud with regard to their technical aspects and differentiations among them.

Cloud computing contains a number of elements which are layered from top to bottom. Bottom layer acts as a foundation for cloud computing. It is a purely physical layer that is involved with providing storage and infrastructure. Top layers are usually involved with the networks and applications and are only possible due to the layers beneath it.

There are three major layers in the cloud i.e., IaaS, PaaS, and SaaS. IaaS lies at the foundation of the cloud paradigm while the SaaS is the layer at the top [17]. Each higher layer incorporates the layers beneath it. Figure 1.5 highlights different layers present in cloud computing. Table 1.2 represents different Layers of cloud computing along with services offered.

Figure 1.5 Layers of cloud computing

Infrastructure as a Service

This is the base layer of the cloud paradigm. This layer consists of physical hardware that is necessary for cloud computing to be effectively operational. This layer contains hardware, network equipment, and web hosting services that can be rented out to the users using the cloud.

The users who want to run their own Cloud-based services can even use this layer.

The users even have options to buy or lease this layer for the above-mentioned purpose. Cloud is based on physical components that can be used for computing purposes. These physical systems include servers, hypervisors, cooling gear etc., these resources are stored in a data center (DC) owned by cloud providers. Since this layer is the base physical layer, without this layer SaaS and PaaS are not possible at all.

Characteristics of IaaS:

Resources shared as a facility.

Active, on-request scaling of services.

Value-based assessing model.

Simultaneous users on a single portion of the hardware.

Platform as a Service

The programmers, coders and the web developers use this layer to develop applications, create programs, software, and web tools [2]. The hardware needed for development purposes is rented out from an IaaS provider. Then, developers can use this layer as a platform to build innovative applications. Examples of infrastructure providers are RackSpace, Amazon EC2, Digital Ocean, etc., Generally, these infrastructures contain pre-installed developer tools. These additional tools can be software libraries that are responsible for performing a specific task or they can be frameworks (such as .Net) or they can even be an application stack such as the LAMP. The LAMP is an application stack that stands for Linux Apache MySQL Perl/PHP since it is hosted on a Linux server and it supports languages such as Perl and PHP.

Characteristics of PaaS:

Distinct environment to code, test, deploy, and repair applications.

UI designing tools to generate, adjust, and deploy diverse UI/UX situations.

Multi-tenant design simplifying simultaneous users.

Load corresponding, safety, and failover abilities for application to be organized.

Dedicated tools for billing and support.

Software as a service

SaaS is the layer above the PaaS layer. This layer is utilized mainly by end-users [18].

It is built on the top of PaaS and IaaS and encompasses both of them to work. It acts as a provider of applications, programs, and web tools. SaaS offer these services either free or at some nominal price. SaaS is the most accessible layer of cloud computing by the end-user and can be accessed on any portable device ranging from smart phones to Laptops.

SaaS has a large number of applications in the business areas. These include customer management system (CRM), accounting and human resource (HR) management. Salesforce.com is the best example of a SaaS vendor.

Characteristics of SaaS:

Integrated Internet-based access to business and commercial applications.

Complete commercial process fluctuating to cloud giving superior services to client.

No trouble regarding software updates and patches as it is the sole responsibility of the service provider.

Application Programming Interfaces (APIs) permit incorporation with different programs

Table 1.2 Different Layers of cloud computing along with services offered

Cloud is divided into four different types of deployment models on the basis of their services they provide.

Types of the cloud

Private cloud

Community cloud

Public cloud

Hybrid cloud

Public cloud

Public cloud is based on a computing model where the cloud provider makes the resources (such as virtual machines, applications or storage), available to everyone over the Internet. Public cloud saves organizations from spending huge amounts of resources to buy, and manage the application infrastructure along with the additional required hardware. In the public cloud, the cloud provider (like Amazon’s AWS or Microsoft’s Azure) would be held responsible for the installation and maintenances of the system along with its additional software dependencies. If implemented in a proper manner, public cloud can be as secure as the private clouds and even at a much lower cost. Effective security methods can be put in place to ensure data privacy. For e.g. Intrusion detection and prevention system (IDPS) can be used to ensure data security. Some of the examples of public cloud service providers are Salesforce, Hp Cloud services, Amazon Web Services etc. Figure 1.6 depicts the public cloud model with its characteristics.

Characteristics of the public cloud:

Easy implementation: Generally, it is easier to implement for the end users as compared to private cloud. Public cloud service providers make it really easy for the end users to access the cloud-based services and applications. Cloud service providers also give free services for some determined trial period which helps users in making a choice to shift towards cloud computing.

Cost-effective in operations: Most public cloud service providers provide their services to the users for free or they follow a Pay as you go (P-A-Y-G) scheme. With this, the users have to pay only for the services they require and nothing more. This ensures effective utilization of the resources and makes it a great cost-effective option.

Rational levels of safety: To ensure safety of the users’ data present on the cloud, public cloud service providers put some security mechanisms in place. They also use user authentication models to prevent unauthorized access of the data stored on cloud.

Figure 1.6 Public cloud model

Private cloud

Private cloud is a cloud computing model whose scope is limited to a single user or an organization. It involves a secure cloud-based environment and only an authenticated client can operate it. This provides the organization with a greater control over their data and more privacy [21]. A company can have its own private cloud that can provide services over a Wide Area Network (WAN). Private cloud model is depicted in figure 1.7.

Characteristics:

Higher security and privacy: In public cloud, services are provided to the clients via a dedicated private network, which makes the data stored on the cloud very secure, and also defend against all sorts of external interruptions.

Considered for big enterprises and businesses: Most of the big companies would be hesitant in putting their sensitive data on a public platform. Public cloud helps them by providing a platform and services which cannot be accessed by the general public. All these services are provided through a private network which is limited to an enterprise only.

Better regulation over cloud infrastructure.

More control over the data stored on the cloud

Cost and energy efficient.

Improved reliability.

Figure 1.7 Private Cloud

Community cloud

A community cloud is simply a broader version of the public cloud. Community cloud supports a community where their members have a similar requirement of services and their share of concerns. The common interests and concerns can include security methods, data privacy, business models and need of end-users [20]. Community cloud can also be further explained on the basis of its abstraction level. A community cloud is delivered over the Internet, which does not happen with the private cloud. They also generally have an operating expenditure price model. Community cloud model is represented in the figure 1.8.

Characteristics:

Better-quality security and privacy.

Improved scalability and flexibility.

IaaS community cloud shares the same cloud infrastructure.

SaaS community cloud shares the same software.

Figure 1.8 Community cloud model

The following table 1.3 highlights different types of cloud along with their scope.

Table 1.3 Types of cloud deployment models with their scope

Hybrid cloud

Hybrid cloud computing model encapsulates two or more same or different types of the cloud’s together. Hybrid cloud can be a combination of a private cloud and a public cloud, a public cloud and a community cloud, or mix of all of them. Every hybrid cloud model has its unique sets of characteristics and features that serves the needs of end users [22]. A clear representation of the hybrid cloud model is shown in figure 1.9.

Figure 1.9 Hybrid Cloud

Characteristics:

Reasonable Price.

Better security, confidentiality, and consistency.

Easiness of data sharing and partnership.

Table 1.4 highlights the key differences between Public, Private, Community and Hybrid Cloud.

Table 1.4 Difference among public cloud, private cloud, hybrid and community cloud

1.6 Features of Cloud Computing

Cloud computing services continue to have a huge influence on the tech industry because of its new way of computing. In the previous section, we talked about different types of cloud models that are present today for the end users. Public clouds have a wide scope and can be used by anyone. They are the most efficient but a wider access can lead to a compromise in data security. In the public cloud, infrastructure is provided off-site over the Internet while in public cloud, the infrastructure is provided on a private network. This limits the scope of the cloud but also ensures the data of the user remains really secure.

Cloud computing comes with many unique characteristics which make it such a hit in today’s world. In this section, we throw light on features and benefits of cloud computing.

Features of cloud computing are shown in figure 1.10.

Features of cloud computing are:

Flexibility: The number of users using the cloud services remains flexible and can be increased or decreased depending on the current need.

Figure 1.10 Features of cloud computing

Self-service provisioning: Services provided by the cloud can be used to do almost any kind of work.

Pay-per-use: Different abstraction models in cloud computing ensures that the users are paying for the cloud services or platform, only on the basis of the services they are using at that moment.

Accessed remotely: Cloud is essentially location-free. It allows the users to work from anywhere as long they have access to an Internet connection. This allows many employees to work on their comfort from their own homes, which ensures a good work-life balance and an increase in productivity.

Ease in server maintenance: For public cloud, since servers are located off-site, it is the responsibility of the providers to maintain these servers. Users are free of the burden of trying to install the server and updating it as needed.

Efficient backup: Using cloud-based backups, a lot of time is saved. It also decreases the involved costs. Using the cloud also reduces the size of their datacenters (DC). Via cloud-based backups the cost towards acquiring servers and software is reduced.

Cloud-based file sharing: Cloud allows streamline teamwork collaboration since data can be shared, accessed and edited from anywhere in the world. Updates can be seen in real-time by all the teammates. Google Docs is a relevant example of this feature.

Resource pooling: In cloud computing, many resources are brought together to assist larger number of users simultaneously as in the case of public and community cloud. These resources are dynamically allocated and de-allocated on the basis of demand. The system is elastic