CLOUD COMPUTING | CS-334 | UNIT - 2

Cloud Computing Architecture:

Cloud Reference Model, Types of Clouds, Cloud Interoperability &
Standards, Scalability and Fault Tolerance; Cloud Solutions: Cloud
Ecosystem, Cloud Business Process Management, Cloud Service
Management. Cloud Offerings: Cloud Analytics, Virtual Desktop
Infrastructure.

Cloud computing architecture

Cloud computing is a utility-oriented and Internet-centric way of delivering IT services
on demand. As seen in the image below.
Cloud computing architecture includes:
1. IaaS, Infrastructure as a service
2. PaaS, Platform as a service
3. SaaS, Software as a service

Cloud infrastructure can be heterogeneous in nature because a variety of resources, such

as
• Clusters
• Networked PCs,
• Databases
• Cloud application
• Cloud programming tools
• Hosting platforms
• Virtual machines, etc. are used.
From the diagram above, we will discuss about:
NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

1. IaaS
2. PaaS
3. SaaS
4. User applications
5. User-level middleware
6. Core middleware
7. System infrastructure
1. IaaS:
1. IaaS stands for infrastructure as a service.
2. Infrastructure as service or IaaS is the basic layer in the cloud computing
model.
3. IaaS offers servers, network devices, load balancers, databases, Web servers,
etc.
4. IaaS examples can be categorized into two categories
1. IaaS Management layer
2. IaaS Physical infrastructure
5. Some service providers provide both above categories and some provide only
the management layer.
6. The IaaS management layer also required integration with other IaaS solutions
that provide physical infrastructure.
7. The main technology behind IaaS is hardware virtualization.
8. Some examples:
1. Amazon Web Services (AWS),
2. Microsoft Azure,
3. Google Compute Engine (GCE)
2. PaaS:
1. PaaS stands for platform as a service.
2. PaaS provides a computing platform with a programming language execution
environment.
3. PaaS offered to the user is a development platform
4. PaaS solutions generally include the infrastructure as well.
5. PaaS offered only the user-level middleware.
6. Some examples:
o Google App Engine
o Force.com
3. SaaS:
1. SaaS stands for software as a service.
2. Software as a service (SaaS) allows users to connect to and use cloud-
based apps over the Internet.

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

3. SaaS is the service with which end users interact directly.

4. Some examples:
o Gmail
o Google Drive
o Dropbox
o WhatsApp
4. User applications:
1. It includes cloud applications through which end users get interact.
2. There may be different types of user applications, like scientific, gaming,
social, etc.
3. Some of the examples are Gmail, Facebook.com, etc.
5. User-level middleware:
1. It includes a cloud programming environment and tools.
2. There may be different types of programming environments and tools
depending on the user applications.
3. Some examples of user-level middleware are Web 2.0, libraries, and scripting.
6. Core middleware:
1. It includes cloud hosting platforms.
2. It manages quality of service.
3. Execution management.
4. Accounting, metering, etc.
5. Virtual machines are the part of core middleware.
7. System infrastructure:
1. It includes cloud resources.
2. Storage hardware
3. Servers, and databases are part of it.

IaaS | Infrastructure as a Service

1. IaaS stands for infrastructure as a service.
2. Infrastructure as service or IaaS is the basic layer in the cloud computing
model.
3. IaaS offers servers, network devices, load balancers, databases, Web servers,
etc.
4. IaaS delivers customizable infrastructure on demand.
5. IaaS examples can be categorized into two categories
1. IaaS Management layer
2. IaaS Physical infrastructure
6. Some service providers provide both above categories and some provide only
management layer.

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

7. The IaaS management layer also required integration with other IaaS solutions
that provide physical infrastructure.
8. On virtual machines applications are installed and deployed.
9. One of example of a virtual machine is an Oracle VM.
10. Hardware virtualization includes workload partitioning, application isolation,
sandboxing, and hardware tuning.
11. Instead of purchasing users can access these virtual hardware on a pay-per-use
basis.
12. Users can take advantage of the full customization offered by virtualization to
deploy their infrastructure in the cloud.
13. Some virtual machines can be with pre-installed operating systems and other
software.
14. On some virtual machines operating systems and other software can be
installed as per use.
15. Some examples:
1. Amazon Web Services (AWS),
2. Microsoft Azure,
3. Google Compute Engine (GCE)

PaaS | Platform as a service

1. PaaS stands for platform as a service.
2. PaaS provides a computing platform with a programming language execution
environment.
3. PaaS provides a development and deployment platform for running
applications in the cloud.
4. PaaS constitute the middleware on top of which applications are built.
5. Application management is the core functionality of the middleware.
6. PaaS provides run-time environments for the applications.
7. PaaS provides
o Applications deployment
o Configuring application components
o Provisioning and configuring supporting technologies
8. For users PaaS interfaces can be in the form of a Web-based interface or in the
form of programming APIs and libraries.
9. PaaS solutions generally include the infrastructure as well.
10. PaaS offered only the user-level middleware.
11. PaaS classification:
o PaaS-I: Runtime environment with Web-hosted application development
platform. Rapid application prototyping. For example, Force.com which
is a combination of middleware and infrastructure product type.
NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

o PaaS-II: Runtime environment for scaling Web applications. The runtime

could be enhanced by additional components that provide scaling
capabilities. For example Google AppEngine which is a combination of
middleware and infrastructure product type. Appscale is middlware
product type.
o PaaS-III: Middleware and programming model for developing
distributed applications in the cloud. For example Microsoft Azure
which is a combination of middleware and infrastructure product
type. Manjrasoft Aneka is a middleware product type.
12. Some examples:
o Google App Engine
o Force.com

PaaS reference model

Characteristics of PaaS:
1. Runtime framework: The runtime framework executes end-user code
according to the policies set by the user and the provider.
2. Abstraction: PaaS offers a way to deploy and manage applications on the cloud
rather than virtual machines on top of which the IT infrastructure is built and
configured.
3. Automation: PaaS deploys the applications automatically.

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

4. Cloud services: Provide services for the creation, delivery, monitoring

management, and reporting of applications.

SaaS | Software as a service

1. SaaS stands for software as a service.
2. Software as a service (SaaS) allows users to connect to and use cloud-based
apps over the Internet.
3. SaaS is the service with which end users interact directly.
4. It provides a means to free users from complex hardware and software
management.
5. In SaaS customers do not need to purchase the software and require the license.
6. They simply access the application website, enter their credentials and billing
details, and can instantly use the application.
7. Customer can customize their software.
8. Application is available to the customer on demand.
9. SaaS can be considered as a “one-to-many” software delivery model.
10. In SaaS applications are built as per the user needs.
11. From the examples mentioned below we can find why SaaS is considered as
one of many models.
12. Some examples:
o Gmail
o Google Drive
o Dropbox
o WhatsApp
Characteristics of SaaS:
1. The product sold to customers is application access.
2. The application is centrally managed.
3. The service delivered is one-to-many.
4. The service delivered is an integrated solution delivered on the contract, which
means it is provided as promised.

Types of clouds
Clouds constitute the primary outcome of cloud computing.
Clouds build the infrastructure on top of which services are implemented and delivered
to customers.
According to the administrative domain, clouds are classified.
Cloud type identifies the boundaries within which cloud computing services are
implemented.
NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

There are four different types of cloud:

1. Public clouds
2. Private clouds
3. Hybrid or heterogeneous clouds
4. Community clouds
1. Public clouds:
1. The cloud is open to the wider public.
2. In the public cloud the services offered are made available to anyone, from
anywhere, and at any time through the Internet.
3. From a structural point of view they are a distributed system.
4. In public clouds one or more data centers are connected together.
5. On these data centers services are implemented.
6. Customer may require his/her credentials or billing details to access the offered
services.
7. Small enterprises prefer public couds due to its less cost.
8. Public clouds offer renting the infrastructure or subscribing to application
services.
9. Customer can use on demand services.
10. Public cloud keeps monitoring of services used by users to provide billing as
per the uses.
11. Public clouds offer any kind of services like,
o IaaS by Amazon EC2 is a public cloud
o PaaS by Google App Engine is a public cloud
o SaaS by SalesForce.com is a public cloud
12. At the same time large quantity of users can access the public cloud.
2. Private clouds:
1. The cloud is implemented within the private premises of an institution and
generally made accessible to the members of the institution or a subset of
them.
2. When customers privacy in important private clouds are preferable over public
clouds.
3. Instead of pay-as-you-go model as in public cloud, there could be other scheme
in private clouds.
4. In private cloud sensitive information are kept in house.
5. Private cloud provides customer information protection better than public
clouds.
6. Private clouds can be implemented on more heterogeneous hardware.
7. Some of the options available for private deployment of clouds we can
consider are DataSynapse, Zimory Pools, Elastra, and Aneka.

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

8. Data Synapse provides a flexible environment for building private clouds on

top of data centers.
9. Elastra Cloud Server is a platform for easily configuring and deploying
distributed application infrastructures on clouds.
10. Zimory provides a software infrastructure layer that automates the use of
resource pools based on Xen, KVM, and VMware virtualization technologies.
11. Aneka is a software development platform that can be used to deploy a cloud
infrastructure on top of heterogeneous hardware: data centers, clusters, and
desktop grids.
3. Hybrid clouds:
1. Hybrid clouds are a combination of private clouds and public clouds.
2. Private clouds stand alone and are sometimes not scalable so here advantages
of public clouds are taken.
3. Public clouds suffer from security threats and administrative pitfalls so the
advantages of private clouds are taken.
4. When advantages of public clouds and private clouds are taken together that’s
known as hybrid clouds.
5. Hybrid cloud allows the services to be taken from public clouds when needed
and keeps the sensitive information within private clouds.
6. Hybrid clouds use cloud bursting, in which services are taken when required
and released when not in use.
7. Dynamic provisioning refers to the ability to acquire on-demand virtual
machines in order to increase the capability of the resulting distributed system
and then release them.
4. Community clouds:
1. Community clouds are distributed systems created by integrating the services
of different clouds to address the specific needs of an industry, a community,
or a business sector.
2. Sectors for community clouds are as follows:
1. Media industry
2. Healthcare industry
3. Energy abd other core industries
4. Public sector
5. Scientific research
3. Community clouds can provide a shared environment where services can
facilitate business-to-business collaboration.
4. Benefits of the community clouds:
1. Openness
2. Community
3. Gracefull failures
4. Convenience and control

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

5. Environmental sustainability
5. Openness: By removing the dependency on cloud vendors, community clouds
are open systems in which fair competition between different solutions can
happen.
6. Scalable: Scalable because the system can grow simply by expanding its user
base.
7. Gracefull failures: Since there is no single provider or vendor in control of the
infrastructure, there is no single point of failure.
8. Convenience and control: The cloud is shared and owned by the community,
which makes all the decisions through a collective democratic process.
9. Environmental sustainability: Hybrid clouds tend to be more organic by
growing and shrinking in a symbiotic relationship to support the demand of
the community, which in turn sustains it.

Cloud Interoperability and standards

Cloud interoperability refers to the ability of customers to use the same management
tools, server images and other software with a variety of cloud computing providers and
platforms.
Standards are important in cloud computing for a variety of reasons. Standards for
interoperability and data and application portability can ensure an open competitive
market in cloud computing because customers are not locked-in to cloud providers and
can easily transfer data or applications between cloud providers.
Why cloud interoperability and standards?
Vendor lock-in can prevent a customer from switching to another competitor’s solution.
If switching is possible, it happens at considerable conversion cost and requires
significant amounts of time. Switching happen because may be customer wants to find
a more suitable solution for customer needs. Or vendor may not be able to provide the
service required. So, the presence of standards that are actually implemented and
adopted in the cloud computing community gives power for interoperability and then
lessen the risks resulting from vendor lock-in.

Scalability and Fault Tolerance

Cloud Scalability is the ability to scale on-demand the facilities and services as and
when they are required by the user.
Cloud Fault Tolerance is tolerating the faults by the cloud that are done by mistake by
the user.
Here the scaling is beyond the limits, it means we can’t even imagine what will be the
limit.
loud middleware is designed on the principle of scalability along different dimensions
in mind e.g.:- performance, size and load.

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

The cloud middleware manages a huge number of resources and users which depends
on the cloud to obtain that they can’t obtain within the premises without affording the
administrative and maintenance costs.
So in this overall scenario the ability to tolerate failure is normal but sometimes it
becomes more important than providing an efficient & optimized system.
The overall conclusion says that “it is a challenging task for the cloud providers to
develop such high scalable and fault tolerance systems who can get managed and at the
same time they will provide a competitive performance.

Cloud Solutions: Cloud Ecosystem, Cloud Business Process Management, Cloud

Service Management
Cloud Ecosystem
Cloud ecosystem is a term used to describe the complex system of interdependent
components that work together to enable cloud services.
The center of a cloud ecosystem is a public cloud provider. It might be an IaaS provider
such as Amazon Web Services (AWS) or a SaaS vendor such as Salesforce.
There is no vendor lock-in in the cloud ecosystem.
For example, AWS is the center of its own ecosystem, but it’s also a part of the
Salesforce ecosystem. Salesforce runs a number of its services on AWS’s infrastructure.
Cloud Business Process Management
1. Business process management (BPM) is a mature business discipline that has
spawned a number of technologies to support it.
2. Today it is the agile who survive those organizations who are able to adapt to
change, to innovate as well as continuously improve, and to continuously
monitor and analyze the results of these adaptations.
3. In the current web enabled business environment, processes in many cases
depend on the discovery and recognition of components that exist as web
services.
4. The current trend is towards increased emphasis on mobility and collaboration
as essential elements to support the agility and currency of business processes.
5. This means that BPM vendors are increasingly seeking to augment their BPM
packages by incorporating greater Web 2.0 type functionality.
6. Cloud based BPM is one response to these new demands.
7. BPM governs organizations cross functional, customer focused end to end core
business process.

Cloud service management

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

Service management:
1. A system integral of supply chain management that contents actual company
sales and the customer.
2. The goal of service management is to maximize service supply chains.
3. The purpose of service management are to reduce high costs by integrating
products and services and keep inventory levels smaller.
Cloud Service Management:
1. Cloud monitoring and cloud service management tools allow cloud providers
to ensure optimal performance, continuity and efficiency in virtualized, on-
demand environments.
2. The delivery of dynamic, cloud-based infrastructure, platform and application
services doesn’t occur in a vacuum.
3. In addition to best practices for effective administration of all the elements
associated with cloud service delivery, cloud service management and cloud
monitoring tools enable providers to keep up with the continually shifting
capacity demands of a highly-elastic environment.
4. The fig illustrates that service management provides the visibility, control and
automation needed for efficient cloud delivery in both public and private
implementations.

Simplify user interaction with it:

1. The user-friendly self-service accelerates time to value.
2. Service catalogue enables standards which drives consistent service delivery.
Enable policies to lower cost with provisioning:
1. Automatic allocating and de-allocating of resources will make delivery of
services fast.
2. Provisioning policies allow release and reuse of assets.
Increase system admin productivity:
1. Providing the benefits to the broker will probably become a critical success
factor in cloud computing.
2. Due to the growth of service brokerage business will increase the ability of
cloud consumers to use services in a trustworthy manner.
NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

3. These cloud mediators will help companies to choose the right platform,
deploy the apps across multiple clouds.
Following are the opportunities for cloud brokers:
1. Cloud service intermediation: The broker must need to manage the additional
securities or management capabilities over the cloud.
2. Cloud aggregation: It includes the deployment of services over multiple cloud
platforms.
3. The ability to group an application across multiple clouds will become
important i.e., if one service goes down the another can be started.

Cloud Offerings: Cloud Analytics, Virtual Desktop Infrastructure

Cloud analytics
Cloud analytics is a service model in which elements of the data analytics process are
provided through a public or private cloud.
Cloud analytics applications and services are typically offered under a subscription-
based or utility (pay-per-use) pricing model.
Gartner defines the six key elements of analytics as
1. Data sources
2. Data models
3. Processing applications
4. Computing power
5. Analytic models
6. Sharing or storage of results
In Gartner’s view, any analytics initiative “in which one or more of these elements is
implemented in the cloud” qualifies as cloud analytics.
Examples of cloud analytics products and services include
1. Hosted data warehouses
2. Software-as-a-Service business intelligence (SaaS BI)
3. Cloud-based social media analytics
It provides users with a better forecasting technique to analyze and optimize the service
lines and provide a higher level of accuracy.
Cloud Business Analytics Competencies:
1. This cloud analytic service is supported by different type of competency areas.
2. The cloud analytics strategy helps clients to achieve their business objectives
faster with less risk and at a lower cost.
3. Another competency is business intelligence and performance management
that helps to increase performance by providing accurate and on time data
reporting.

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

4. The next is analytics and optimization that provides different types of

modeling techniques, deep computing and simulation techniques to increase
the performance.
5. The other competency is a content management system that includes the
different service architecture, technology architecture.
How Cloud Analytics Work?
1. Cloud Analytics works with the combination of hardware services and
middleware.
2. The system features include the platform that provides data reporting analytics
based on text mining activities and analytics techniques.
3. This system also takes care of storage optimization and different high
performance data warehouse management techniques.
Analytics Business Outcomes:
1. Analytics help to get the right information as and when they are required
identify how to get it and point out the right sources to get it.
2. Analytics also helps in designing the policies faster based on the information
available in the organization.
3. This gives the option through which the organization can increase its profit,
reduce cycle time and reduce defects.

Testing under cloud

Cloud testing typically involves monitoring and reporting on real-world user traffic
conditions as well as load balance and stress testing for a range of simulated usage
conditions.
Load and performance testing conducted on the applications and services provided via
cloud computing particularly the capability to access these services in order to ensure
optimal performance and scalability under a wide variety of conditions.
Consumers can access the IT resources in the test environment.
Testing under the cloud gives very good sign by decreasing the manual intervention and
reducing the processes in the typical testing environment.
After enabling of resources as and when they are required ,it reduces the investment on
capital as well as enables the business to handle the ups and downs of the testing
requirements.

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

Facts under cloud computing

The fig clearly shows that on the basis of these six parameters a cloud testing process
can be performed.
Advantages of Cloud Testing:
1. Reduces capital investment and operational costs and not effect goal critical
production application.
2. Offers new and attractive services to the clients and present an opportunity to
speed cycles of innovations and improve the solution quality.

Virtual Desktop Infrastructure

Virtual desktop infrastructure (VDI) is defined as the hosting of desktop environments
on a central server. It is a form of desktop virtualization, as the specific desktop images
run within virtual machines (VMs) and are delivered to end clients over a network.
Virtual Desktop Infrastructure (VDI) is a concept in which a solution based on a server
based computing model that is not so different from the traditional terminal server
centralized computing model used to deliver applications to remote users.
Virtual Desktop Infrastructure or VDI is the name given to a collection of technologies
and processes that dramatically extend the concept of a remote desktop.
It centers on the idea that companies can virtualized their desktop operating systems
like Windows XP or Vista using platforms like VMware ESX or Xenserver. and run
said desktops from within the secured data center.

NIDA KHAN
Dept of CSE
 CLOUD COMPUTING | CS-334 | UNIT - 2

Vdi Benefits:
1. Data Security
2. Reduced Hardware Expenditures
3. Easier Management: Perform centralized patching and application installation
or streaming without loading the network
4. Mobile Workforce: User desktops are portable – users can reconnect from any
location with a variety of devices.
5. Resource Pooling.

NIDA KHAN
Dept of CSE