Mr. M.

Madhusudhan
[email protected]

CLOUD COMPUTING
 Cloud Computing
Unit II
Foundations (Part – I)
T1: Ch1, Ch2, Ch3, Ch4
Course Objective: Outline Cloud Architecture and its
models
Course Outcome: Understand Cloud architecture and
its models
Common Terminology

2
 Definition of Cloud

• Cloud computing is using the internet to access

someone else's software running on someone
else's hardware in someone else's data center.
Lewis Cunningham

3
 Definition of Cloud…

• Cloud computing is a style of computing where

massively scalable IT-related capabilities are
provided ‘as a service’ across the Internet to
multiple external customers.

Gartner

4
 Definition of Cloud…

• A large-scale distributed computing paradigm that

is driven by economies of scale, in which a pool of
abstracted, virtualized, dynamically scalable,
managed computing power, storage, platforms,
and services are delivered on demand to external
customers over the Internet.
Ian Foster

5
 Definition of Cloud…
• A Cloud is a type of parallel and distributed system
consisting of a collection of interconnected and
virtualized computers that are dynamically
provisioned and presented as one or more unified
computing resources based on service-level
agreements established through negotiation
between the service provider and consumers.
Rajkumar Buyya

6
 Definition of Cloud…
• Berkeley summarized the key characteristics of
cloud computing as:
(1) the illusion of infinite computing resources;
(2) the elimination of an up-front commitment
by cloud users; and
(3) the ability to pay for use . . as needed . .

7
 Definition of Cloud…
• NIST characterizes cloud computing as
pay-per-use model for enabling
available,
convenient,
on-demand network access
to a shared pool of configurable computing
resources

8
Roots of Cloud Computing

Hardware

Distributed Cloud Internet

Computing
Computing Technologies

Systems
Management

9
 Roots of Cloud Computing…
• SOA
• Web Services
• Web 2.0
• Grid Computing
• Utility Computing
• Autonomic Computing

10
 Layers/Services of Cloud
• Web Browser (Access Management Tool)
• Cloud Applications
SaaS • Ex: Social Networks, Office suites, CRM

• Cloud Development Environment (Access Management Tool)

• Cloud Platform
PaaS • Ex: Programming Languages, Frameworks, Structured Data, Mashups editors

• Virtual Infrastructure Manager (Access Management Tool)

• Cloud Infrastructure
IaaS • Ex: Compute Servers, Data Storage, Firewall, Load Balancer

11
 Desired Features of a Cloud
• Self-Service
• Per-Usage Metering and Billing
• Elasticity
• Customizationo

12
 Motivation for Cloud Computing
Startups have many reasons for moving from an intranet-
based, capital-purchase dependent model of IT
infrastructure to a utility-style demand and cloud-based
service. These include the following:
• Scalability – Cloud computing helps startups manage
shifting computing requirements by providing greater
flexibility in the computing services they purchase. A
cloud-based IT infrastructure is more versatile – notably
in terms of scalability – than is local, intranet-based
infrastructure.
13
 Motivation for Cloud Computing
• Reliability – Because cloud vendors can build more redundancy
into a system than a company can build into its own intranet, the
cloud vendor can spread its infrastructure investment costs across
its entire customer base, allocating resources as necessary.
• Virtualization – Because cloud-based IT infrastructure can be
virtualized and geographically dislocated, startups are freed from
having to consider the physical location of its IT infrastructure and
data centers in business operations decisions.

14
 Motivation for Cloud Computing
• Affordability – Under traditional infrastructures, startups may
not receive – or have financial wherewithal to purchase –
certain features that are often offered to cloud computing
customers at substantial discounts. How do these benefits
pass on to startups and other small companies? Because the
marginal cost to the cloud computing provider of many
features (such as enhanced security) may be very low (or even
negligible), otherwise unaffordable services may be offered for
free to startups using cloud computing options

15
 Cloud Computing IS a Platform

Cloud computing applications develops by leveraging

platforms and frameworks. Various types of services are
provided from the bare metal infrastructure to customize-able
applications serving specific purposes.

16
 Cloud Computing IS a Platform
• Amazon Web Services (AWS) –
AWS provides different wide-ranging clouds
IaaS services, which ranges from virtual
compute, storage, and networking to complete
computing stacks. AWS is well known for its
storage and compute on demand services,
named as Elastic Compute Cloud (EC2) and
Simple Storage Service (S3).

17
 Cloud Computing IS a Platform
• Google AppEngine –
Google AppEngine is a scalable runtime environment
frequently dedicated to executing web applications.
These utilize benefits of the large computing
infrastructure of Google to dynamically scale as per the
demand. AppEngine offers both a secure execution
environment and a collection of which simplifies the
development if scalable and high-performance Web
applications.

18
 Cloud Computing IS a Platform
• Microsoft Azure –
Microsoft Azure is a Cloud operating
system and a platform in which user can
develop the applications in the cloud.
Generally, a scalable runtime
environment for web applications and
distributed applications is provided.

19
 Cloud Computing IS a Platform
• Hadoop –
Apache Hadoop is an open source framework
that is appropriate for processing large data sets
on commodity hardware. Hadoop is an
implementation of MapReduce, an application
programming model which is developed by
Google. This model provides two fundamental
operations for data processing: map and
reduce. 
20
 Principles of Cloud Computing
The six principles - detailed in the recently published ISACA
publication 
Guiding Principles for Cloud Computing Adoption and Use - include
enablement, cost/benefit, enterprise risk, capability, accountability
and trust. Here's how ISACA defines each of those principles:
1.Enablement: Plan for cloud computing as a
strategic enabler, rather than as an outsourcing
arrangement or technical platform.

21
 Principles of Cloud Computing
2.Cost/benefit: Evaluate the benefits of cloud acquisition based on a
full understanding of the costs of cloud compared with the costs of
other technology platform business solutions.
3.Enterprise risk: Take an enterprise risk management perspective to
manage the adoption and use of cloud.
4. Capability: Integrate the full extent of capabilities that cloud
providers offer with internal resources to provide a comprehensive
technical support and delivery solution.

22
 Principles of Cloud Computing
5.Accountability: Manage accountabilities by
clearly defining internal and provider
responsibilities.
6.Trust: Make trust an essential part of cloud
solutions, building trust into all business processes
that depend on cloud computing.

23
 Five Essential Characteristics
1.On-demand self-service: A consumer can unilaterally
provision computing capabilities, such as server time and
network storage, as needed automatically without
requiring human interaction with each service provider.
2. Broad network access: Capabilities are available over
the network and accessed through standard mechanisms
that promote use by heterogeneous thin or thick client
platforms (e.g., mobile phones, tablets, laptops and
workstations).
24
 Five Essential Characteristics
3.Resource pooling: The provider's computing resources are pooled
to serve multiple consumers using a multi-tenant model, with
different physical and virtual resources dynamically assigned and
reassigned according to consumer demand. There is a sense of
location independence in that the customer generally has no control
or knowledge over the exact location of the provided resources but
may be able to specify location at a higher level of abstraction (e.g.,
country, state or datacenter). Examples of resources include storage,
processing, memory and network bandwidth.

25
 Five Essential Characteristics
4.Rapid Elasticity: Capabilities can be elastically provisioned and
released, in some cases automatically, to scale rapidly outward and
inward commensurate with demand. To the consumer, the
capabilities available for provisioning often appear to be unlimited
and can be appropriated in any quantity at any time.
5. Measured Service: Cloud systems automatically control and
optimize resource use by leveraging a metering capability at some
level of abstraction appropriate to the type of service (e.g., storage,
processing, bandwidth and active user accounts). Resource usage
can be monitored, controlled and reported, providing transparency
for the provider and consumer.
26
 Cloud Deployment Models
The four deployment models:
These are:
• Public Cloud Model
• Private Cloud Model
• Hybrid Cloud Model
• Community Cloud Model

27
Cloud Deployment Models

28
 Cloud Deployment Models
1.Public Cloud Model:
• The name speaks for itself: public clouds are available to the
general public, and data are created and stored on third-party
servers.
• Server infrastructure belongs to service providers that manage
them and administer pool resources, which is why there is no
need for user companies to buy and maintain their hardware.
Provider companies offer resources as a service both free of
charge or on a pay-per-use basis via the Internet connection.
Users can scale resources when required.
29
 Cloud Deployment Models
• The public cloud deployment model is the first choice
for businesses that operate within the industries with
low privacy concerns. When it comes to popular
public cloud deployment models, examples are 
Amazon Elastic Compute Cloud (Amazon EC2) — the 
top service provider, Microsoft Azure, Google App
Engine, IBM Cloud, Salesforce Heroku and others.

30
Cloud Deployment Models

31
 Cloud Deployment Models
The Advantages of a Public Cloud
1.Hassle-free infrastructure management:Having a third party running your
cloud infrastructure is convenient: you do not need to develop and maintain
your software as the service provider does it for you. Also, the infrastructure
setup and use are unsophisticated.
2.High scalability:You can easily extend the available capacity as your
company requirements increase.
3.Reduced costs:You pay only for the service you use, no need to invest in
hardware or software.
4.24/7 uptime: An extensive network of the provider’s servers ensures the
constant availability of your infrastructure and its improved operation time.

32
 Cloud Deployment Models
The Disadvantages of a Public Cloud:
1.Compromised reliability: That same server network is also meant to ensure
against failures. But often enough, public clouds experience outages and
malfunction, as it was in the case of the Salesforce CRM disruption in 2016
that caused a storage collapse.
2.Data security and privacy give rise to concern: Although access to data is
easy, a public deployment model deprives users of knowing where their
information is kept and who has access to it.
3.The lack of a bespoke service: Service providers have only standardized
service options, which is why they will probably fail to satisfy their
requirements if they are unusual.

33
 Cloud Deployment Models
Private Cloud Model:
• There is little to no difference between a public and a private model from
the technical point of view, as their architectures are very similar. However,
opposed to a public cloud that is available to the general public, only one
specific company owns a private one. That is why it is also called
an internal or corporate.
• Only one organization uses this deployment model to run its workloads,
and the server can be hosted externally or on the premises of the user
company. Regardless of their physical location, these infrastructures are
maintained on a designated private network and use software and
hardware that are intended for a specific company.

34
 Cloud Deployment Models
• A clearly defined scope of persons have access to the information kept in a
private repository, which prevents the general public from using it. In light
of numerous breaches, a growing number of large corporations decided on
a closed private type as it is expected to be less risky.
• If compared to a public cloud, a private one provides wider opportunities
for customizing the infrastructure to the company’s requirements. A
private model is especially suitable for companies that seek to safeguard
their mission-critical operations or for businesses with changing
requirements.
• Multiple public cloud service providers — including Amazon, IBM, Cisco,
Dell and Red Hat — also provide private solutions along with public ones.

35
Cloud Deployment Models

36
 Cloud Deployment Models
The Benefits of a Private Cloud
1.All the benefits of this deployment model result from
its autonomy. They are the following
2.Bespoke and flexible development and high scalability,
which allows companies to customize their
infrastructures in accordance with their requirements
3.High security, privacy and reliability as only authorized
persons can access resources
37
 Cloud Deployment Models
The Drawbacks of a Private Cloud
• The major disadvantage of the private cloud
deployment model is its cost, as it requires
considerable expenses on hardware, software and
staff training. That is why this secure and flexible
computing deployment model is not a choice for
small companies.

38
 Cloud Deployment Models
3.Community Cloud Model:
• A community deployment model largely resembles a private one; the only
difference is the set of users. While a private type implies that only one
company owns the server, in the case of a community one, several
organizations with similar backgrounds share the infrastructure and related
resources.
• If the organizations have uniform security, privacy and performance
requirements, this multi-tenant data center architecture helps companies
achieve their business-specific objectives. That is why a community model is
particularly suited for organizations that work on joint projects. In that case,
a centralized cloud facilitates project development, management and
implementation. Also, the costs are shared across all users.
39
Cloud Deployment Models

40
 Cloud Deployment Models
The Strengths of a Community Cloud:
• Cost reduction
• Improved security, privacy and reliability
• Ease of data sharing and collaboration
The Shortcomings of a Community Cloud:
• High cost if compared to a public deployment model
• Sharing of fixed storage and bandwidth capacity
• It is not widespread so far
41
 Cloud Deployment Models
4.Hybrid Cloud Model:
• As it is usually the case with any hybrid phenomenon, a hybrid cloud
encompasses the best features of the above-mentioned deployment
models — public, private and community ones. It allows companies
to mix and match the facets of all three types that best suit their
requirements.
• As an example, a company can balance its load by locating mission-
critical workloads on a secure private cloud and deploying less
sensitive ones to a public one. Hybrid cloud deployment model not
only safeguards and controls strategically important assets but does
so in the most cost- and resource-effective way possible for each
specific case. Also, this approach facilitates 42
Cloud Deployment Models

43
 Cloud Deployment Models
The Benefits of a Hybrid Cloud:
• Improved security and privacy
• Enhanced scalability and flexibility
• Reasonable price
However, the hybrid deployment model only makes
sense if companies can split their data into mission-
critical and non-sensitive.

44
The Comparison of Top Cloud Deployment Models
Public Private Community

Ease of setup and use Easy Requires IT proficiency Requires IT proficiency Requires IT proficiency

Data security and privacy Low High Comparatively high High

Data control Little to none High Comparatively high Comparatively high

Reliability Vulnerable High Comparatively high High

Scalability and flexibility High High Fixed capacity High

Cost-intensive, the most expensive Cheaper than a private model but more costly
Cost-effectiveness The cheapest one Cost is shared among community members
one than a public one

Demand for in-house

No Depends Depends Depends
hardware

45