Cloud Concepts And Models Tutorial

1.1 Cloud Concepts and Models.

Hello and welcome to the module of Cloud Concepts and Models of the CompTIA Cloud Plus course offered by Simplilearn. This module introduces you to cloud computing from the National Institute of Standards and Technologies (NIST) perspective. This module also explains different models of cloud computing. Let us discuss the objectives of this module in the next slide.

1.2 Objectives

By the end of this module, you will be able to: Define cloud computing Explain the essential characteristics of cloud computing Describe the service models of cloud computing Describe the deployment models of cloud computing Identify the accountability and responsibility in case of outage Explain the object storage concepts Let us begin with the definition of cloud computing, in the next slide.

1.3 Definition of Cloud Computing

According to NIST, “Cloud computing is a model for enabling ubiquitous, 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.” Let us look at the definition again with our focus on the key words this time. The first key word is Ubiquitous. It means irrespective of any type of network or web enabled device, customer can access the services subscribed from the service provider. The second key word is Convenient. Cloud computing provides similar technology, services, and applications as the internet, but with a self-service utility. The third key word is on-demand network access. According to the latest version of the NIST definition, cloud computing networks must use virtualization and multi-tenancy. Finally the key words rapidly provisioned and released, and minimal management effort or service provider interaction mean that the customer is able to provision the resources with as little management effort and service provider interaction as possible. Let us move on to an interaction, which depicts the different characteristics of cloud in the next slide.

1.4 Cloud Computing Characteristics

NIST states that there are five defining characteristics to be realized by service providers to be able to call their service a cloud service. The five defining characteristics are On-demand Self-Service, Broad Network Access, Resource Pooling, Rapid Elasticity, and Measured Service. Click each button for details of the characteristic. On-demand self-service means a client can obtain, use, and remove cloud resources without any assistance from the service provider. It provisions the resource needed as per the project requirement. The customer gets the complete privilege to select the resources and decide its quantity based on their business process without any formal consultation with the service provider. The customer can Upgrade/downgrade the subscribed resources anytime without service provider’s interaction. The customer can access their dedicated portal anytime and from anywhere. However, it should be in compliance with the limitation of regulatory concern to perform changes in configuration of current setup. When resources are hosted either in an organization’s private cloud network, public network, or hybrid deployment, they are available for access from a range of devices. This accessibility of the resources is called Broad Network Access. This service is available everyday throughout the year. The range of devices can include tablets, smartphones, Windows PCs, Mac PCs, etc. This feature also allows online access to resources from different locations. In public cloud computing, Internet is the backbone, as without it, the public cloud service would not be available. In case of private cloud, the communication and interaction take place in the LAN environment. Broad Network Access lets the employee access from any web enabled device. This is the reason why the demand for Broad Network Access exists as more employees use their personal smartphones or tablets to work and want online access to company resources extended to these devices. This service is accessible via any type of network. The concept of resource pooling refers to the grouping together of resources (assets, equipment, personnel, effort, etc.) for the purpose of maximizing advantage and/or minimizing risk to the users. In short, it refers to the scalability of the physical infrastructure. If there are more customers, the service provider can scale up the hardware resources without introducing a downtime to the existing customers. Due to resource pooling, we achieve multi-tenancy. Multi-tenancy is adopted by cloud due to several reasons. The table on the slide shows the advantages of multi-tenancy over single tenancy. In single-tenant architecture, the software instance is used by a single customer only. This customer may be allowed to access and customize the application’s code. In multi-tenant architecture, the tenants refer to the multiple customers using a single software instance. Multi-tenancy can be more economical than single-tenancy as the software development and maintenance costs are shared. In the case of software upgrades, in multi-tenant architecture, all customers are upgraded with a new code given at the same time, minimizing the service provider’s time and effort spent on upgrades. In single-tenancy, customers have to be upgraded separately, one at a time. Rapid deployment is the combination of scalability and flexibility. Scalability allows the service providers and customers to scale up or scale down the resource based on their customer’s strength. Flexibility allows the consumers to scale up or down the resources based on their company requirement and client load using their dedicated control panel. In cloud computing, there is minimal interaction with the service provider which gives total control of the service usage to the customer. Cloud computing provides a great advantage of unlimited resources which is achieved by the service provider using a concept called dynamic provisioning. This will be covered in later modules. Due to this characteristic, customer can have total control over the subscribed system, i.e., on-demand self-service is achieved. Cloud computing follows the pay-as-you-go model. This model states that the customer will be charged based on their usage and not based on rental. Hence, customer will have the financial control of the services. Measured service means log activity generation, monitoring, and maintenance. Thus, it can be said, due to measured service, the financial flow of the infrastructure can be maintained based on the number of customer threshold. In the next slide, we will look into an interaction to explain the different service models of cloud computing.

1.5 Different Service Models

The different service models of cloud computing are: Infrastructure as a Service Platform as a Service Software as a Service Communications as a Service Data as a Service Business Process as a Service X as a Service Click each tab for the details of the service model. Infrastructure as a service is generally used by administrators, architects, and operators. IaaS provides virtual hardware resources on-demand. It provides multiple on-demand features like individual, email, and domain name servers; messaging systems; private networks; etc. IaaS applications may incur OS license fees and may require installation of compatible software on the servers. Here, the customer has the flexibility to provision and de-provision the resources as per their business demand. The image on the slide shows the responsibility matrix where the part marked in green is taken care by cloud service provider and the remaining by the cloud customer. An entire computing infrastructure can be built by an organization using this application. This type of service is beneficial for start-up companies. It will help the company focus on its business process as IT infrastructure management and uptime are maintained by the IaaS cloud service provider. Some examples of Infrastructure as a Service are Rackspace, VMware VCloud, etc. Platform as a Service was introduced for Developers. It is required to run the software products, which requires servers (physical), web servers, and database software. The application is active on software stack along with compilers, dependency files, etc. The disadvantage of Platform as a Service is building the platform for the application to run on is a lengthy task, and has to be updated and monitored on a regular basis. The image on the slide shows the responsibility matrix where the part marked in green is taken care by cloud service provider and the remaining by the cloud customer. Platform as a Service is a cloud service platform for the developers, which provides them a development and deployment platform. Development platform contains all tools and dependency files required to develop new software by the developer. Deployment platform, however, contains all controls to maintain the customers of that developer for whom the product is to be deployed. Platform as a Service provides an external platform to execute software applications without any administrative requirements of the lower level components. Some examples of Platform as a Service are Microsoft Windows Azure, force.com, etc. Software as a Service is an application generally used by the end-user. This service is on a pay-as-you-go basis, and is supplied through a network as and when required. The advantage of this model is its lack of installation requirement. A network connection and a browser complete the requirements for Software as a Service. The software is installed in the service provider’s application server and the customer gets the software service off-shore, i.e., software delivered from the server which is outside organization. Here, since this is in cloud, we can say multi-tenancy is also achieved by the service provider. The image on the slide shows the responsibility matrix where the part marked in green is taken care by cloud service provider and the remaining by the cloud customer. Although, the service provider has only one copy of installation, multiple users can be handled with their dedicated storage space. Some of the examples of Software as a Service are Microsoft Office Live, Oracle on Demand, and salesforce.com. Communications as a Service is generally leased from a single vendor. It is an enterprise communication solution that is outsourced, which refers to the services hosted outside an organization. Communications as a Service can contain Voice over IP or VoIP, collaboration, applications using mobile and fixed devices to videoconference, and Instant Messaging (IM). Communications as a Service vendor assures guaranteed Quality of Service (QoS), and is in charge of all hardware and software management. Businesses operating on Communications as a Service can deploy communications devices and modes as and when required on a pay-per-use basis. The advantage of Communications as a Service is, it does not require large capital investment and don’t incur ongoing overhead. With Communications as a Service, small and medium-sized businesses can be more flexible and expandable. It allows advantages like on-demand coverage, and addition of devices or modes. The network capacity and features are changed daily if necessary. This is done to avoid resource wastage, and to ensure the functionality are at par with the demand. This ensures the system is not outdated and would not require significant replacements or upgrades. Google Talk embedded in Gmail Service, WhatsApp, BBM, etc. are some examples of Communications as a Service. Data as a Service is a model for distribution and information provision. Through this model customers have access to various data files via a network like internet. Data files include text, images, sounds, and videos. Through this model, customers and client-oriented enterprises are offered solutions which are both convenient and cost-effective. Data as a Service permits, but does not need, data cost and usage to be separated from platform or software cost and usage. The two types of pricing offered by Data as a Service vendors are volume-based and format-based. Volume-based pricing contains a fixed cost per megabyte of data in the whole repository. In format-based pricing the charge is fixed based on the format of the data. Data as a Service is easily able to move data between platforms. This avoids the conflict and confusion arising due to multiple copies of the same data or file. Data as a Service implements access control measures like strong encryption and passwords, avoiding ‘vendor lock-in’, ease of administration and collaboration, diverse platform compatibility, automatic updates, and global accessibility. These measures preserve data integrity. Any horizontal or vertical business process delivered based on the cloud service model is known as Business Process as a Service. Software as a Service, Platform as a Service, and Infrastructure as a Service are reliant on this service. With the advent of cloud computing, a more service-oriented approach is preferred by companies. Instead of assuming that a packaged application that includes business logic, data, and processes is needed, it is possible to select a process application that is not tied to a single application. A business is unable to forecast the future leverage of a business process, hence, a Business Process as a Service must support multiple languages and deployment environments. A Business Process as a Service environment must be able to handle massive scaling. A service must be able to progress from handling a few to an increasing number of processes and customers. The service accomplishes that objective by optimizing the underlying cloud services to support this type of elasticity and scaling. X as a Service denotes the growing number of services distributed over the internet instead of local or on-site provision. It is at the core of cloud computing. X as a Service uses hybrid cloud computing to deliver IT as a service. It refers to either a single or a blend of Software as a Service, Infrastructure as a Service, Platform as a Service, Communications as a Service, or Business Process as a Service. X as a Service is a term used frequently for the services on private or public clouds which were previously detached and are now integrating and becoming transparent. Let us look into the different deployment models in cloud computing in the next slide.

1.6 Different Deployment Models

Public Cloud, Private Cloud, Community Cloud, and Hybrid Cloud are the four main deployment models of cloud recommended by the National Institute of Standards and Technology (NIST). Applying cloud can decrease the expense for any business, but, it is also associated with challenges and security risks for IT management. These can be expensive for an organization to deal with. To reduce the risk, it is essential for a business to recognize their requirements prior to opting for the different existing deployment models on the cloud. Let us understand the first deployment model, public cloud, in the following slide.

1.7 Public Cloud

Public cloud is a model wherein public and organizations can access the cloud infrastructure, however, the owner is a third-party cloud service provider. The real concept of cloud hosting is represented by the public cloud deployment model. Google is an example of this model. This model provides many clients with the required services and infrastructure. This model is available on a pay-per-user license policy or free-of-charge basis depending on the vendor. The public cloud deployment model is suitable for business requirements. The reason is, in business, load spike management is a necessity. Load spike is a term used in software. The different characteristics of load spikes are balancing request for network connection, hosting SaaS applications, and utilizing the available temporary infrastructures to develop and test applications, and managing user-consumed applications. If the user-consumed applications are not managed, large investments in infrastructure would be required from businesses. CAPital EXpenditure or CAPEX and OPerational EXpenditure or OPEX is decreased using this model. We will move on to the concept of private cloud in the following slide.

1.8 Private Cloud

In this model, hosting is built and maintained specifically for each client, which ensures data security. The necessary infrastructure can be on-site or at a third-party location. Private Cloud can be a cloud residing within an organization, or it can be hosted in another data-center leasing organization. If it is hosted in another data-center leasing organizations or hosted by cloud provider, it is termed as Virtual Private Cloud or VPC. Private cloud is not cost efficient, however, the advantage of this model is the level of security it offers. When adopting the cloud, data security becomes a point of concern for many organizations. A secure-access VPN or the physical location within a client’s firewall system addresses the security concerns. This model is also adopted by organizations which need applications or data to adhere to various regulatory standards like SOX, HIPAA, or SAS70. These standards mandate the management of data, for privacy and audits that govern the corporation. Many SaaS applications like SugarCRM give their customers choice where they can keep the data on their premise to assure data privacy. Another choice available is of virtual private cloud which is offered by Amazon. The major advantage of Private cloud is the total cost of ownership with respect to hardware and other components remains with the organization. Community cloud is the next deployment model dealt with in the next slide.

1.9 Community Cloud

In this model, many organizations share the infrastructure offered by cloud. The policy and compliance considerations are the same throughout. This requires various organizations to co-operate. This model is shared by a bigger group, which decreases the cost further in comparison to private cloud. A community cloud is used for management of data and applications. This is beneficial for many government departments on a state-level who need to access the same data. Let us focus on the last deployment model, hybrid cloud in the next slide.

1.10 Hybrid Cloud

This model enables different businesses to utilize the secured applications and data hosting available on a private cloud. However, businesses continue to get cost benefits as the shared applications and data are kept on a public cloud. This model has benefits of both private and public cloud. If the present private cloud infrastructure is unable to manage the load spikes, and needs a back-up to support the load, this is called cloud bursting, which is also managed by this model. A migration of workloads between private and public clouds is assisted without inconveniencing the user. Application Programming Interfaces or APIs are exposed by several PaaS deployments. This can be combined with the internal or private cloud hosted applications without compromising on the security features. The two examples of this model are force.com and Microsoft Azure. Hybrid cloud is more secure since here the customer can maintain highly sensitive data with their servers and less sensitive data with the cloud service provider’s server. Let us now look at the different categories of accountability in the next slide.

1.11 Network Accountability

The characteristics of a cloud complicates the process of accountability. The three categories of accountability in case of an outage, identified by the telecommunications industry are: product-attributable service outages, customer or service-provider attributable outages, and external attributable outage. Product-attributable service outages are the failures in software or hardware, outages attributed to customers or service providers, force majeure-attributable or external outages like natural disasters or some malicious acts. They are mainly off-set by a system design, hardware, software, components, or other parts of the system. The design of the system makes scheduled outages a necessity. Customer or service provider attributable outages are mainly offset by errors in the procedure, environment of the office, etc. The outages caused by natural disaster or third parties are known as external attributable disaster. Examples of natural disaster are floods, earthquakes, etc. The third-party caused outages may or may not be related with customer or supplier. In the next slide, let us now find out the different responsible parties of the cloud models.

1.12 Network Responsibility

In cloud the following parties are responsible: Cloud consumers need to correctly provision, configure, and operate their application. They are held accountable in these roles. Virtual appliance suppliers are responsible for producing only software applications. They are accountable to ensure these software-only applications are dependable and stable. Infrastructure suppliers provide hardware and software for the platform in which the cloud service provider operates. They also supply computers, storage, and networking equipment. They need to assure their devices are dependable and robust. Cloud service providers are held accountable for providing dependable and robust operation in the infrastructure and facilities of cloud computing. They also should ensure that the needs of the cloud consumers are met. Network service providers need to ensure the smooth functioning of the network and make sure that the network is continuously available. End users are also held responsible if there is a fault in the operation, configuration, device, or equipment failure at their premise. Let us spot the differences between on-premises and off-premises services in the next slide.

1.13 Different Services Comparison

The table on the slide depicts the difference between on- and off-premises services. The first point cost refers to the setting up of an environment for a business process based on business needs. The second point refers to the time required to bring the setup into production mode. Both cost and time for on-premises setup is more than off-premises. The third point is troubleshooting, this deals with repairing unsuccessful processes or products and is considered to be a form of problem solving. The fourth point availability refers to the uptime of services and data. The responsibilities of both troubleshooting and availability for on-premises deployment model lies with the customer, whereas, for off-premises, the service provider is responsible. Let us consider the fifth point now which is security. Security refers to the three important points: confidentiality, availability, and integrity. It is the responsibility of the customer to maintain security in on-premises service and due to this; the security is more. In the case of off-premises service, responsibilities are distributed according to the SLA. The sixth point is hardware ownership. The complete hardware and its access are with the customer, in the case of on-premises setup, whereas it is with the service provider in the case of off-premises service. This seventh point is data ownership. Data ownership refers to who owns the data and who can access the data. For on-premises setup, it is the customer who owns the data, whereas, in the case of off-premises setup, the data ownership depends on the SLA. We will move on to the important cloud concepts in the next slide.

1.14 Cloud Concepts

The important cloud concepts discussed in detail here are: IT Chargeback, Cloud Bursting, Orchestration Platforms, and Object Storage. Let us begin with the first concept chargeback in the next slide.

1.15 IT Chargeback

IT chargeback is defined as the concept of monitoring IT expenses of different business units of various IT organizations and accordingly charging them back. In other words, IT chargeback refers to tracking and measuring the IT expenses for each business unit within an organization, so this information can be included in budget estimations. Chargeback applies to cloud computing in a similar way. When an organization uses a private cloud, each business unit within the organization uses it to a different extent. Whether the cloud service is provided internally by the IT department or externally by a service provider, it becomes important to track usage by department or function so that the costs can be calculated and charged back to each department’s budget. By tracking private cloud usage, business units within an organization can become more sensitive to and responsible for the resources they use. Today, there are a number of third-party packages available to track private cloud usage, although an organization may choose to build cloud accounting tools in-house. In the next slide, we will look into the concept of cloud bursting.

1.16 Cloud Bursting

When a company uses up its internal computing resources, it then shifts its extra workload to an external cloud. This is done on an on-demand basis and the process of shifting the workload is called ‘bursting’ and the entire concept is known as cloud bursting. The external cloud is called the public cloud and the company has to pay for it on a pay-as-you-go basis. The internal computing resources of the company is referred to as a private cloud. In a hybrid cloud, the deployment is able to cloudburst, in other words, it is able to distribute the workload to the public cloud. Cloud bursting is not the only advantage of hybrid clouds. They also assist in load balancing. However, factors such as performance, security, availability of enterprise-class service, etc. have to be taken into consideration before one deploys a hybrid cloud. This helps an organization to be responsible in their usage and duties with respect to IT computing. Due to cloud bursting, the company can track and measure the IT expenses. In the next slide, we will discuss orchestration platforms.

1.17 Orchestration Platforms

Orchestration Platform is an essential component for an IT administrator, which provides an automated way to manage the cloud setup in an organization. For example, with an orchestration platform, a developer would request creation of VM via the portal dedicated for him which provides self service functions. Orchestration platforms can also be used for centralized management solutions in a private or public cloud. Orchestration is similar to automation in which all functions related to admin tasks are automated by a single click of button. Let us move on to the concept of object storage, in the following slide.

1.18 Object Storage

Object storage is a redundant and highly scalable cloud storage service that allows users to easily store, search, and retrieve data across the Internet with optional Content Delivery Network (CDN) connectivity. In the OSD standard, objects are termed as a flexible data container which is dynamically scaled based on usage. In the OSD standard, objects are specified with a 64-bit partition ID and a 64-bit object ID. In the OSD standard, the term metadata refers to "data about data". The term is ambiguous, as it is used for two fundamentally different concepts. Structural metadata is about the design and specification of data structures and is more properly called "data about the containers of data". Descriptive metadata, is about individual instances of application data, the data content. In the OSD standard, Binary Large Objects (BLOB) is a collection of binary data stored as a single entity in a database management system. Blobs are typically images, audio, or other multimedia objects though sometimes binary executable code is stored as a blob. Let us move on to look into some case studies in the following slides.

1.19 Case Studies

The case studies discussed are: Facebook and Akamai Mahindra Satyam and Amazon Indiagames.com and Amazon Redbus.in and Amazon Let us move on to the first case study, Facebook and Akamai in the following slide.

1.20 Case Studies Facebook and Akamai

Facebook is a social utility that connects people with friends; and others who work, study, and live around them. However, to maintain the infrastructure uptime, high availability, and big data; it relies on a content delivery solution provider AKAMAI Technologies. When Facebook started using AKAMAI servers to store garbage data, they found that the availability zones were increased, which resulted in less network latency and high availability. Here, Facebook refers garbage data as the images and the multimedia clips uploaded by the user. Availability zones are the servers who are responsible for providing faster access and communication to the data. Facebook and Akamai together form hybrid cloud deployment model, but the deployment model between Facebook and their end users is community cloud deployment model. In the next slide, we will discuss the case study of Mahindra Satyam and Amazon.

1.21 Case Studies Mahindra Satyam and Amazon

Mahindra Satyam needed a way to share and manage documents for one of its clients. The client was one of the top three global pharmaceutical companies. Mahindra Satyam turned to Amazon Web Services (AWS) for a solution. Using various products and services, AWS delivers a secure, reliable, and effective result to Mahindra Satyam. The various products and services include Amazon Elastic Compute Cloud (Amazon EC2), AWS, etc. AWS provide a virtual infrastructure with good availability zone for sharing and managing documents. Mahindra enjoyed business profit through resource sharing. In the next slide, we will discuss the case study of Indiagames.com and Amazon.

1.22 Case Studies Indiagames com and Amazon

Indiagames Ltd. is India's benchmark mobile and online games company and a leading global mobile game publisher. Switching from its own infrastructure to the scalable infrastructure of Amazon Web Services (AWS) enabled the company to reduce the time to market for their games from weeks to days, make decisions quickly, and benefit from a variable cost structure. Indiagames Ltd. totally focused on its actual business process without worrying about the IT infrastructure since the responsibility of uptime was committed by Amazon Web Services. In the next slide, we will discuss the case study of Redbus.in and Amazon.

1.25 Summary

Here is a quick recap of what was covered in the module: IT chargeback refers to tracking and measuring the IT expenses for each business unit within an organization. There are five essential characteristics of cloud. They are: on-demand self-service, rapid elasticity, measured service, resource pooling, and broad network access. There are seven service models of cloud computing. They are: Software as a Service, Platform as a Service, Infrastructure as a Service, Communications as a Service, Data as a Service, Business Process as a Service, and X as a Service. There are four deployment models of cloud. They are: private, public, hybrid, and community cloud.

1.26 Thank You

In the next module, we will discuss virtualization.

  • Disclaimer
  • PMP, PMI, PMBOK, CAPM, PgMP, PfMP, ACP, PBA, RMP, SP, and OPM3 are registered marks of the Project Management Institute, Inc.

We use cookies on this site for functional and analytical purposes. By using the site, you agree to be cookied and to our Terms of Use. Find out more

Request more information

For individuals
For business
Name*
Email*
Phone Number*
Your Message (Optional)

By proceeding, you agree to our Terms of Use and Privacy Policy

We are looking into your query.
Our consultants will get in touch with you soon.

A Simplilearn representative will get back to you in one business day.

First Name*
Last Name*
Email*
Phone Number*
Company*
Job Title*

By proceeding, you agree to our Terms of Use and Privacy Policy