Chapter 3

Cloud Computing Architecture

© Oxford University Press 2018. All rights reserved

 Learning Outcomes

At the end of the session you will be able to:

• Comprehend characteristics of cloud computing
• Define grid computing
• Explain important features of cloud and grid computing
• Describe grid architecture and cloud computing architecture
• Differentiate between grid and cloud computing

© Oxford University Press 2018. All rights reserved

 Introduction

• Grid computing is the integration of computer resources for achieving similar

objective.
• The grid may be a dispersed system along with non-interactive workloads
which comprise a huge number of files.
• Grids are frequently created with middleware software libraries of a common
grid.

© Oxford University Press 2018. All rights reserved

 Grid Framework

• Grids are a type of dispersed computing system, whereas a virtualized super

computer is made from various networked.
• Grids focus on two different but associated objectives—supplying isolated
access to IT resources and building up processing control.
• The grid is a technology which controls two factors—allocation and trust.
• Grid computing is a versatile technology which has its base in e-science and
has progressed from previous expansions in parallel and high-performance
computing (HPC).

© Oxford University Press 2018. All rights reserved

 Grid Framework

The main resources that can be shared in a grid are:

1. Processing and computing power
2. Networked file and data storage systems
3. Bandwidth and communications
4. Application software
5. Tools used for scientific purpose

© Oxford University Press 2018. All rights reserved

 Grid Framework

The distinct definitions for grid and grid computing are:

• Grid middleware is exclusive software that offers the essential functionality
needed to facilitate sharing of various resources and setting up of virtual
businesses. Grid middleware is exclusive software that is incorporated into
the infrastructure of the concerned corporation. Grid middleware offers a
unique virtualization and sharing layer which is positioned among the various
infrastructures and the particular user applications using it.
• Grid computing is fundamentally the installed Grid middleware or the
computing permitted by grid middleware based on synchronized, safe,
flexible resource sharing among a collection of resources, people, and
organizations. © Oxford University Press 2018. All rights reserved
 Grid Framework

• Grid infrastructure refers to the union of grid middleware and hardware

which converts single portions of data resources and hardware into an
incorporated virtualized infrastructure that is displayed to the user as the
only computer in spite of heterogeneity of the fundamental infrastructure.
• Utility computing is a type of computing that provides customized
applications of grid and computing as a service. It is based on
pay-as-per-utilization business modules.

© Oxford University Press 2018. All rights reserved

 Grid Architecture

The grid design offers an outline of the grid constituents, describes the
objective and operations of its constituents, and shows how the constituents
interrelate with each other. Various layers of the grid architecture are:
• Fabric layer
• Connectivity layer
• Resource layer
• Collective layer
• Application layer

© Oxford University Press 2018. All rights reserved

 Grid Architecture
The key functionalities of a grid middleware are as follows:
1. Integration and virtualization of various independent resources
2. Requirement of information concerning resources and their accessibility
3. Lively and flexible resource administration and allotment
4. Brokerage of resources based on open markets or corporation strategies
5. Safety comprising agreement and confirmation of users and accountability
6. Licenses administration
7. Expense and invoicing
8. Transport of non-insignificant ‘Quality of Service’
© Oxford University Press 2018. All rights reserved
 Challenges of Grid Computing
• It is not enough to just change the prevailing spread IT infrastructure into a grid.
In majority of instances, investments are required for making the existing
applications to work on a grid infrastructure.
• Lack of values for grid computing make resources’ findings for grid technology
difficult and risky.
• Grid computing is a versatile technology and the launching of grid computing in a
corporation is characteristically a long-standing plan that needs time until the
visibility of first results. The beginning of grid computing could need consistency
of physical resources. Even if grids would essentially be capable to handle
heterogeneity of resources that are accessible, advanced heterogeneity of
resources might need advanced savings in terms of money and time and hence
increase the downfall risk.
© Oxford University Press 2018. All rights reserved
 Cloud Computing Architecture

• In cloud computing environment, physical resources are made available to

cloud users with the help of virtualization software in the form of virtual
resources.
• Cloud computing architecture is not fixed as other computing architectures,
but it is different on the basis of different jobs, resource distribution.

© Oxford University Press 2018. All rights reserved

 Cloud Computing Architecture
Basic cloud computing
environment is shown in the Fig.
3.1 in which various services such
as available servers, virtual
desktop, system software,
application software, database, etc.
are available for cloud users. Users
can avail the services using any
device such as desktop system,
laptop, mobile, tablet, etc.,
© Oxford University Press 2018. All rights reserved
 Cloud Computing Architecture

The Cloud Computing architecture is discussed based on different criteria:

• On the Basis of Load Balancing
• On the Basis of Disk Provisioning
• On the Basis of Storage Management
• On the Basis of Hypervisor Installed
• On the Basis of Migration
• On the Basis of Service Relocation
• On the Basis of Cloud Balancing
• On the Basis of Virtual Switches Load Balancing
• On the Basis of Failure Detection and Recovery

© Oxford University Press 2018. All rights reserved

 Key Design Aspects of Cloud Architecture, Cloud Services and
Cloud Applications

• Cloud computing has various issues related to its architecture:

• Issues at Design Level
Issues of Architectural of Cloud Computing
Platform Related Issues
• Issues Related to the Implementation
Issues related to Business
Technical Issues

© Oxford University Press 2018. All rights reserved

Similarities and Differences between Grid and Cloud Computing
Similarities:
• Grid computing and cloud computing are scalable.
• Grid and cloud computing offer service-level agreements (SLAs) for uptime
accessibility of around 99 per cent. If the service slides below the definite uptime
service level, the service credit for receiving data to the customer gets delayed.
Differences:
• Cloud computing offers SaaS applications, whereas grid computing is used when the
processing authority of service or an application is distributed across multiple
systems.
• With cloud computing, the corporation acquires feasibility and expense savings,
whereas with grid computing the corporation acquires suppleness and authority.
© Oxford University Press 2018. All rights reserved
 Features of Cloud Computing
• On-demand self service
• Broad network access
• Resource pooling
• Rapid elasticity
• Measured service
• Multi-persistence
• Dynamic Computing Infrastructure
• Elasticity and scalability
• IT service-centric approach
• Standardized Interfaces © Oxford University Press 2018. All rights reserved
 Cloud and Dynamic Infrastructure
• Cloud service characteristics include reduced cost, high scalability,
pay-as-you-grow facility, guaranteed data center and network uptime, better
resource utilization, and lesser implementation time since resources are
available on demand.
• Cloud services are exclusive as they permit to see, amend and share files saved
on the cloud. At the same time, as no two cloud services are alike, all of the
services offer almost similar basic functionality and features.
• The finest cloud computing services are those which permit to store and
upload any sort of file one would keep on his/her local hard drive, from word
documents to music files. Security issues can be solved by file encryption, and
password sharing is restricted to protect the data.
© Oxford University Press 2018. All rights reserved
 Cloud and Dynamic Infrastructure
Points to consider while selecting a cloud provider as per the requirement are
given below:
• Ease of use
• Help and support
• Packaged software
• Shared software
• Tool management
• Mobile access
• Line of business services
• Customer-oriented services © Oxford University Press 2018. All rights reserved
 Impediments to Cloud Adoption

Businesses must vigilantly think about five chief barriers for prosperous
implementation of cloud computing. These are as follows:
• Security
• Privacy
• Irresponsibility of vendors
• Risk improvement
• Inheritance of applications

© Oxford University Press 2018. All rights reserved

Thank You!

© Oxford University Press 2018. All rights reserved