CORE COURSE- XI :

23PCA11- CLOUD COMPUTING

Unit-II

Dr. A. Kangaiammal, MCA, M.Phil., M.E., Ph.D.

AP/Computer Applications
Government Arts College(Autonomous)
Salem – 636 007.
 COURSE OBJECTIVES
At the end of this the learners will :
– Understand the current trends and basics of Cloud
computing.
– Know the types of virtualization.
– Understand Big Data Analytics principles and
techniques.
– Understand the financial value of Big Data Analytics.
– Explore the tools and practices for working with Big
Data.
– Emphasize the use of Hadoop, MapReduce.
 COURSE OUTCOMES
On successful completion of the course, the
students will be able to:
– Collaborate the cloud services to any device.
– Implement virtualization.
– Identify the need of big data analytics for any
domain.
– Use Hadoop, MapReduce Framework.
– Apply big data analytics for a given real-time
problem
REFERENCE BOOKS:
1. V.K. Pachghare, “Cloud Computing”, PHI, 2012.
2. Seema Acharya, Subhasini Chellappan, “Big Data and Analytics”, Wiley
Publication, 2017.
3. BIG DATA, Black Book TM, DreamTech Press, 2015.
4. Michael Miller, Cloud Computing: Web–Based Applications That Change
the Way You Work and Collaborate Online, Que Publishing, August 2008.
5. Frank J Ohlhorst, “Big Data Analytics: Turning Big Data into Big Money”,
Wiley and SAS Business Series, 2012.
6. Anand Rajaraman and Jeffrey David Ullman, Mining of Massive Datasets,
Cambridge University Press, 2012.

WEB RESOURCES:
7. https://azure.microsoft.com/en-in/overview/what-is-cloud-computing/ .
8. https://www.ibm.com/cloud-computing/learn-more/what-is-cloud-compu
ting/

9. https://www.salesforce.com/in/cloudcomputing/
 Cloud Computing Architecture
• There are many components in the architecture of
Cloud computing.
• All these components are loosely connected with each
other.
• Cloud architecture can be classified as:
– Front end
• Client/user side consists of computer, different application and
interfaces needed to access the cloud computing platforms.
– Back end
• Includes all resources like computers, servers, storage devices,
deployment models, virtual machines, services and security
mechanisms.
• Provides traffic control, built-in security and different protocols.
• It consists of servers with different protocols with device
interfaces.
 Client/User Interface

Internet

Service Servers

Infrastructure
Storage Applications

Security
 Cloud Computing Models
• Models are classified into two types:
– Service Models
– Deployment Models
• Service models
– SaaS Models
– PaaS Models
– IaaS Models
– HaaS Models
 Cloud Service Model
S
a
a On-demand applications e.g. salseForce.com

P S
a Development platforms, scaling,
a middleware, OS eg. Cloudbees

S
Pay by usage, memory, disk,
network. Eg. Rackspace
 Service Models
User

SaaS

PaaS

IaaS

Physical Infrastructure
 SaaS
• Cloud Computing infrastructure to be shared
by multiple users for one application
irrespective of physical location.
• Capability provided by the service provider to
user to access the applications available on
the cloud infrastructure.
• SaaS model signifies a hosted application
available over a network via internet.
• Client can access the applications through a
thin client interface i.e. browser.
 SaaS
• It supports SOA and Web Services
• Pay-per-user model
• Offer services at low cost
• Benefits from commercial software without
installation at the local.
• Concurrent users at a time.
• Using browsers accessible; security to
browsers is very important.
• Encryption, SSL and web services security
 SaaS Architecture
Client
Client

SaaS SaaS
Provider Provider

SaaS SaaS
Provider Provider

Client Client
 Characteristics of SaaS
• Software management and access through
network.
• Different activities are controlled from SP’s
locations.
• Users use any applications anywhere using internet.
• One-to-many model instead of one-to-one.
• Feature updation done centrally; so users need not
download upgrades or patches.
• Users may request additional features, no cost
• New features released often as whole group is
benefited.
 SaaS Services
1. Release of simple applications to
users
2. Release of single application or
group of software to business
3. Release of group of applications
related to a particular business
4. Desktop as a Service (DaaS)
 Advantages of SaaS
1. Reduced Time
2. Every element is a service
3. Ideal for agile software development
4. Reduced total cost and decrease in the
operational cost
5. Scalability and Integration
6. Availability of upgrades
7. Easy-to-use
8. Security
 Disadvantages of SaaS
1. Robustness 3. Security
2. Privacy 4. Reliability

Security issues in SaaS

• Identity management
• Cloud Standards are weak
• Secrecy
 Vulnerabilities in SaaS Model
1. Multiple users of the same
infrastructure
2. The unknown data location to the user
3. Difficulties to provide security in web
application
4. Attackers productive major target
5. Virtualization Vulnerabilities
 Platform as a Service (PaaS)
• Builds an environment which supports
development of applications.
• Variety of software uploaded and available on the
server with CSP.
• Automates configuration, deployment and
management of applications.
• Complex application can be created – Ex. ERP
• Rental services may be OS, storage, hardware and
network capacity
• Developers get appropriate OS, middleware,
databases, managed services and software tools
 PaaS
• Provides more than IaaS
• VM works as a channel, but machines face
problem from attackers.
• Maintaining integrity of applications is very
important.
• PaaS reduces the overhead cost to almost
one-third.
• Very difficult to write and migrate applications
• Surmount issues of managing various tools
 Characteristics of PaaS
• Development environment is browser-based.
• Scalability, built-in security, and interfaces for
web services.
• Tools for defining business rules, approval
processes and workflow.
• In the same platform, it supports for
integration with other applications
• For outside applications, it provides interfaces
for web services interfaces.
 Requirements of a PaaS Platform
1. High Scalability and on-demand
provisioning
2. High Availability
3. High reliability
4. Optimal usage
5. Auto-scaling
6. Administration/management console
reports
7. Multi-OS and multi-language support
 Issues in PaaS
1. Lack of portability between PaaS clouds
2. Event-based processor scheduling
3. Security Engineering

Benefits of PaaS
1. Reduce the administrative overhead
2. Cost of ownership is less
3. Scalable solutions
4. Latest versions of software
 Drawbacks of PaaS
1. Data Security
2. Limited flexibility
3. Customer captivity
4. Problems of Integration
 Factors for the Selection of PaaS
1. Size of the organization
2. Resources required for maintenance of the
services and processes
3. Resources required for maintenance of the
infrastructure
4. Flexibility
5. Requirements about security and compliance
6. Security and cost facators
- Public PasS offers low cost per service, scalability
and availability
- Private PasS offers security and performance
 Types of PaaS based on deployment
method
1. Public - small orgns, flexibility, no need
to manage infra but limited db,
languages with less flexibility.
2. Private – protected by firewall, privacy
and control, hence security
3. Hybrid – private security and public
deployment and testsing
 Types of PaaS based on the functions
performed

1. Standalone development
environments
2. Application delivery-only
environments
3. Open platform as a service
4. Add-on development facilities
 PaaS Solutions
- Offers complete development tools
1. Social application development
platforms
2. Web-based application add-on
platforms
3. Standalone business application and
platforms
4. Open computing platforms
 Vulnerabilities in PaaS
1. CSP accountable for the security and privacy
of user’s data; users can use additional
security mechanism as per the requirements.
2. Not only with web-applications but SOAs too.

Main PaaS Provider

1. Appistry
2. Heoku
3. Windows Azure
4. Force.com
5. Google App Engine
 Infrastructure as a Service (IaaS)
• Responsible for h/w and bare min s/w for users.
• Offers basic services like virtual servers, data
storage and database.
• Users use these for deploy and run applications
• IaaS provider – Rackspace
• IaaS provides foundation for PaaS and SaaS.
• IaaS aims to provide standard, flexible and virtual
environment.
• Supports for access, monitor and manage remote
infrastructures. (pay-on-use basis)
 IaaS
• IaaS providers are:
– Microsoft Azure
– Amazon Web Service (AWS)
– Google Compute Engine (GCE)
• Key concepts of IaaS:
– Cloud Bursting – software for reallocation
processes –client-specific proprietary software.
– Multi-tenant Computing – pay-as-you-go on large
scale applications, servers, storage, computing –
multi-tenancy achieved using virtual machines.
 IaaS Architecture

Server

Server

Server
Firewall
Load
User
Balancer

VM Deployment

VM Automation
Owner of
Software VM

VM uploaded to storage
VM

Iaas Service Provider

Storage
 Characteristics of IaaS
• VMs have preinstalled software
• Resources are distributed as a service
• Resources are available on-demand
• Dynamic scaling of computing resources is allowed
• Cost is based on use of the service, cost is variable
• IaaS supports mulitple users to use a single piece
of hardware
• IaaS supports to accumulate the copies of a
specific data at various locations.
 Advantages of IaaS
• Reduction of capital expenditure on HW and
infrastructure.
• No need of high skilled manpower for the management
of hw and infra
• Reduction of risk in ROI.
• Upgradation of s/w in infra hired (also db, tools and infra
of the user) is user responsibility
• Pricing model not affordable-install on dedicated server
• Multiple platforms –different development options
• Better security than existing software
• New software releases, Data centre and backup are
taken care by IaaS providers.
• CSP accountable for failover of data centre or software.
 Disadvantages of IaaS
• Success of use depends on CSP capability.
• Long-term service use may be costly.
• Upgradation of s/w developed is user
responsibility.
• Changing the existing user software as per
hardware is a binding.
• Prevention of off-premise and out-of-country
data storage due to legal reasons.
• CSP may offer inadequate security features.
• High speed internet may not be offered by CSP.
Benefits of IaaS

Benefits
of IaaS
 Issues of IaaS
• Network dependent and web browsers
related issues.
• Compatibility with legacy security
vulnerabilities
• Virtual machine collapse
• Strength of virtual machine level
isolation
• Data erase practices
 IaaS Classifications
• Compute as a service
• Web hosting
• Storage as a service
• Disaster recovery and backup as a service
• Desktop as a service
• Servers as a service
• Networking as a service
 Vulnerabilities in IaaS
• Security related to OS, applications and
data is user responsibility.
• Public cloud has more risk chances.
• Third-party infra for transmission is risky.
• Exact locations of data unknown
• After use of VMs for execution, data may
reside in storage devices.
 Main IaaS Providers
• Amazon
• Joyent
• Layered Technologies
• Terremark
• GoGrid
• Rackspace
• AT & T
• VMware
 Comparisons of Service Models
S. Model Consumer Service Service Customization
No. Offered Coverage
1 SaaS End User Finished Application Minimum
application uptime and customization
performance depends on the
provider or market
2 PsaaS Owner of the Runtime Environment High degree of
application environment for availability, customization within
application performance constraints of the
code, storage, without service offered, many
other services application applications will need
such as coverage to be rewritten
integration

3 IaaS Owner of the Virtual server Virtual server Minimal customization

application or IT cloud storage availability, time based on constraints
provides OS to provision, no on applications
middleware and platform or installed on
app support application standardized virtual
coverage OS builds
 Deployment Models
• Public Cloud
• Private Cloud
• Community Cloud
• Hybrid Cloud
 Public Cloud
• Services owned an operated by CSPs
• Accessible to all users
• Resources dynamically allotted
• Infra owned by SPs; public cloud owned by third-
party vendor; Services available in common infra;
multiple users use; infra location unaware;
• Less secure than other models
• Limited configuration of infra with less securty and
availability of resources
• Ex. Social networking and email.
 Public Cloud Trade-offs
• Utility model
• No agreement
• Hardware sharing
• No control on performance
• Examples of public cloud:
• Google
• Amazon AWS
• Microsoft
 Public Cloud is useful when
• Large no. of users require to use standardized workload
for applications. Ex. Email
• User want to develop and test some applications
• User wants to check the maximum capacity of the
computing resources for peak times.
• For some projects in collaboration with others
Disadvantages of Public Cloud
• Public cloud is more vulnerable than private
• It is less adaptive than private cloud
Advantages of Public Cloud

Adva
ntage
s
 Private Cloud
• Dedicated use of a particular organization
• Set-up using data-centre of an organization
• Resources are scaled and pooled through virtual
applications by vendor
• Users can share and use these resources from
private cloud.
• Handles the issues on security and privacy of
data.
• Expensive than public cloud.
Private Cloud Types and trade-offs
• Two types:
– On-premise private cloud
– Externally hosted private cloud
• Trade-offs in private cloud:
– Security
– Compliance
– Customizable
– Hybrid deployments
 Private Cloud is useful when
• Private cloud is useful if:
– Cloud efficiency and data independence are
needed
– Consistency across services is required
– Data centre of organization becomes more
efficient
– More server capacity than the requirement of
organization.
 Advantages
• Private cloud ensures more security and
privacy
• Has ore control on its infrastructure and
resources
• More efficient
• More reliable than public cloud
 Disadvantages
• Services are provided only locally not globally
• Costly as hardware cost is more
• Scaling is allowed only within the limit of
internal resources hosted by a system.
• For management and maintenance, skilled
persons required.
• For deployment of cloud, skilled experts are
required.
 Community Cloud
• Many organizations together implement a cloud model.
• It is a multi-tenant model.
• Cloud is specifically used by specific users from the
organizations.
• Users have shared concerns and have similar
requirements such as mission, security, etc.
• Infra in one or all premises of organizations or off all
premises.
• Community cloud is public cloud with additional level of
security, privacy and policy considerations.
 When is the Community Cloud useful?
• Within a state with different government organizations that require to
share their resources
• For a group of hospitals.

Advantages
• Lower cost
• Share resources among involved organizations
• Secure than public cloud

Disadvantages
• Privacy is a major issue.
• Issues on the responsibilities of governance, cost and security
Architecture of community cloud model

Cloud

Organization 1 Organization 2 Organization 3

User 1 User 2 User 3 User 4 User 5 User 6

 Hybrid Cloud
• Private cloud linked to multiple external cloud(s),
two or more cloud models linked together.
• Benefits of multiple deployment models like
private, public and community models.
• When private runs out of capacity, hybrid is
needed.
• Infra managed centrally and acts as a unique
entities.
• More secure control of data and applications
• But to keep track of multiple cloud platforms
 When is Hybrid cloud useful?
• Organization requires to use an application
run on a SaaS but security is the concern
• Organization provides services that are
tailored as per growth of the markets. Public
for interacting with clients and a private for
secure storage of data.
• One can use private cloud for internal use,
while public cloud for customers.
 Advantages
• Provides both public and private cloud
scalability.
• Both secure resources and scalable public
resources.
• More cost effective as it is both public and
private clouds.
• More security
 Disadvantages
• Formed by combination of more than one
cloud, networking is more complex.
• Security policies of the organization are the
major issue.
• Based on the internal infrastructure
• Assurance of redundancy across the cloud
data-centres.
 Comparison of Public, Private and Hybrid
clouds
Public Cloud Private Cloud Hybrid Cloud
The services provided by a The infrastructure and the It is a combination of
cloud are owned and entire cloud is managed public and private clouds
operated by a cloud either by the organization
provider or by a third party.
Lower cost agility pay for More expensive Lower costs
service

Less secure as compared to More secure than public More secure control of the
other cloud models cloud model data and applications

Provide services for Cloud infrastructure is -

enterprises operated exclusively for
some specific organization
 Identity as a Service (IDAAS)
• Information related with a digital media stores
in a form which can be used in electronic
communications.
• Major functions of identity services are
Storage of data, Query and Policy engine
which maintain integrity of data.
• Important for better trust.
• The servers those run the various Internet
domains are IDaaS servers.
 Identity
• A set of characteristics which gives something with
a recognizable output is identity.
• An identity of a person has following things:
– Things you are
– Things you know
– Things you have
– Things you relate to
• To establish the identity of a network:
– One-factor authentication
– Two-factor authentication
– Multi-factor authentication
 IDaaS
• IDaaS may include following things:
– Identity governance
– Authentication services
– Risk and event monitoring
– Identity and profile management
– Single sign-on services
– Directory services
– Policies, roles and enforcement
– Federated identity
– External policy administration
– Registration
Thank you