CLOUD COMPUTING

1. Introduction

Cloud computing is Internet-based computing, whereby shared resources, software, and

information are provided to computers and other devices on demand, like the electricity
grid.

Cloud computing is a paradigm shift following the shift from mainframe to client–server
in the early 1980s. Details are abstracted from the users, who no longer have need for
expertise in, or control over, the technology infrastructure "in the cloud" that supports
them.

1
 CLOUD COMPUTING

2. What Is Cloud Computing?

Cloud computing is Internet based computing where virtual shared servers provide
software, infrastructure, platform, devices and other resources and hosting to customers on
a pay-as-you-use basis. All information that a digitized system has to offer is provided as a
service in the cloud computing model.

Users can access these services available on the "Internet cloud" without having any
previous know-how on managing the resources involved. Thus, users can concentrate more
on their core business processes rather than spending time and gaining knowledge on
resources needed to manage their business processes.

Cloud computing customers do not own the physical infrastructure; rather they rent the
usage from a third-party provider. This helps them to avoid huge. They consume resources
as a service and pay only for resources that they use. Most cloud computing infrastructures
consist of services delivered through common centers and built on servers.

Sharing resources amongst can improve, as servers are not unnecessarily left idle, which
can reduce costs significantly while increasing the speed of application development.

2
 CLOUD COMPUTING

3. History of Cloud Computing

Concept originated from telecommunication companies changing to VPN

 1999: Salesforce.com ‐ Delivery of applications via web

 2002: Amazon launches Amazon Web Services (AWS)

 2006: Google Docs, Amazon Elastic Compute Cloud (EC2)

 2008: Eucalyptus

 2009: Microsoft Azure

3
 CLOUD COMPUTING

4. Architecture of Cloud Computing

Cloud architecture, the systems architecture of the software systems involved in the
delivery of cloud computing, typically involves multiple cloud components
communicating with each other over application programming interfaces, usually web
services.

Cloud computing sample architecture

This resembles the UNIX philosophy of having multiple programs each doing one thing
well and working together over universal interfaces.
Complexity is controlled and the resulting systems are more manageable than their
monolithic counterparts.

The two most significant components of cloud computing architecture are known as the
front end and the back end. The front end is the part seen by the client, i.e. the computer
user. This includes the client’s network (or computer) and the applications used to access
the cloud via a user interface such as a web browser.

The back end of the cloud computing architecture is the ‘cloud’ itself, comprising various
computers, servers and data storage devices.

4
 CLOUD COMPUTING

5. Types of Cloud Computing

Public cloud

Public clouds are made available to the general public by a service provider who hosts the
cloud infrastructure. Generally, public cloud providers like Amazon AWS, Microsoft and
Google own and operate the infrastructure and offer access over the Internet. With this
model, customers have no visibility or control over where the infrastructure is located. It is
important to note that all customers on public clouds share the same infrastructure pool
with limited configuration, security protections and availability variances.

Public Cloud customers benefit from economies of scale, because infrastructure costs are
spread across all users, allowing each individual client to operate on a low-cost, “pay-as-
you-go” model. Another advantage of public cloud infrastructures is that they are typically
larger in scale than an in-house enterprise cloud, which provides clients with seamless, on-
demand scalability. These clouds offer the greatest level of efficiency in shared resources;
however, they are also more vulnerable than private clouds.
A public cloud is the obvious choice when:

o Your standardized workload for applications is used by lots of people,

such as e-mail.
o You need to test and develop application code.
o You need incremental capacity (the ability to add compute resources
for peak times).

o You’re doing collaboration projects.

5
 CLOUD COMPUTING

Private cloud

Private cloud is cloud infrastructure dedicated to a particular organization. Private clouds

allow businesses to host applications in the cloud, while addressing concerns regarding data
security and control, which is often lacking in a public cloud environment. It is not shared
with other organizations, whether managed internally or by a third-party, and it can be
hosted internally or externally.

There are two variations of private clouds:

1. On-Premise Private Cloud: This type of cloud is hosted within an organization’s

own facility A businesses IT department would incur the capital and operational
costs for the physical resources with this model. On-Premise Private Clouds are
best used for applications that require complete control and configurability of the
infrastructure and security
2. Externally Hosted Private Cloud: Externally Hosted Private Cloud are also
exclusively used by one organization but are hosted by third party specializing in
cloud infrastructure. The service provider facilitates an exclusive cloud
environment with full guarantee of privacy. This format is recommended for
organizations that prefer not to use a public cloud infrastructure due to the risks
associated with the sharing of physical resources.

6
 CLOUD COMPUTING

Undertaking a private cloud project requires a significant level and degree of engagement to
virtualize the business environment, and it will require the organization to reevaluate
decisions about existing resources. Private clouds are more expensive but also more secure
when compared to public clouds. An Info-Tech survey shows that 76% of IT decision-
makers will focus exclusively on the private cloud, as these clouds offer the greatest level of
security and control.

When is a Private Cloud for you?

o You need data sovereignty but want cloud efficiencies

o You want consistency across services
o You have more server capacity than your organization can use
o Your data center must become more efficient
o You want to provide private cloud services

Hybrid cloud

Hybrid Clouds are a composition of two or more clouds (private, community or public) that
remain unique entities but are bound together offering the advantages of multiple
deployment models. In a hybrid cloud, you can leverage third party cloud providers in

7
 CLOUD COMPUTING

either a full or partial manner; increasing the flexibility of computing. Augmenting a

traditional private cloud with the resources of a public cloud can be used to manage any
unexpected surges in workload.

Hybrid cloud architecture requires both on-premise resources and off-site server based
cloud infrastructure. By spreading things out over a hybrid cloud, you keep each aspect of
your business in the most efficient environment possible. The downside is that you have to
keep track of multiple cloud security platforms and ensure that all aspects of your business
can communicate with each other.

Here are a couple of situations where a hybrid environment is best:

o Your company wants to use a SaaS application but is concerned about

security.
o Your company offers services that are tailored for different vertical
markets. You can use a public cloud to interact with the clients but keep
their data secured within a private cloud.
o You can provide public cloud to your customers while using a private
cloud for internal IT.

Cloud Stakeholders

To know why cloud computing is used let’s first concentrate on who use it. And then
we would discuss what advantages they get using cloud. There are three types of
stakeholders cloud providers, cloud users and the end users [Figure 1]. Cloud providers

8
 CLOUD COMPUTING

provide cloud services to the cloud users. These cloud services are of the form of
utility computing i.e. the cloud users uses these services pay-as-you-go model. The
cloud users develop their product using these services and deliver the product to the end
users.

Interconnection between cloud stakeholders

9
 CLOUD COMPUTING

6. Components of cloud computing:

Here the components in the sense refer to the platforms like cloud delivery, usage of the
network’s front end back end which together form the cloud computing architecture.
1. Storage-as-a-service: In this component, we can avail of storage as we do it at the
remote site. It is the main component and called disk space on demand.
2. Database-as-a-service: This acts as a live database and the main aim of this
component is to reduce the price of dB by using more software and hardware.
3. Information-as-a-service: Data that can approach from anywhere is known as
information-as-a-service. Internet banking, online news, and much more are included
in it.
4. Process-as-a-service: Combination of different sources like information and services
is done in process-as-a-service; it is mainly helpful for mobile networks.
5. Application-as-a-service: It is a complete application that is ready to use and it is the
final front end for the users. Few sample applications are Gmail, Google calendar, and
much more.
6. Platform-as-a-service: In this component, the application is being advanced and a
database is being advanced.
7. Integration-as-a-service: This deals with components of an application that are built
and need to integrate with other applications.
8. Security-as-a-service: This component is required by many customers because
security has the initial preference.
9. Management-as-a-service: This component is useful for the management of clouds.
10. Testing-as-a-service: This component refers to the testing of applications that are
hosted remotely.
11. Infrastructure-as-a-service: This component takes all the virtual hardware, software,
and servers, networks. It also prevents the running of multiple servers.

10
CLOUD COMPUTING

11
 CLOUD COMPUTING

7. Recent Development’s of Cloud Computing

In 2007, Google, IBM, and a number of universities embarked on a large scale cloud
computing research project. In early 2008, Eucalyptus became the first open source AWS
API compatible platform for deploying private clouds.

By mid-2008, Gartner saw an opportunity for cloud computing "to shape the relationship
among consumers of IT services, those who use IT services and those who sell them”, and
observed that "organizations are switching from company-owned hardware and software
assets to per-use service-based models" so that the "projected shift to cloud computing ...
will result in dramatic growth in IT products in some areas and significant reductions in
other areas.”

In March 2010, Microsoft's CEO, Steve Ballmer, made his strongest statement of betting
the company's future in the cloud by proclaiming "For the cloud, we're all in" and further
stating "About 75 percent of our folks are doing entirely cloud based or entirely cloud
inspired, a year from now that will be 90 percent."
Hewlett Packard’s chief technology and chief strategy officer Shane Robison in July 2010
named the “inevitability of the cloud” as a top tenet of HP’s strategy. Cloud computing is
one of eight key research focus areas for HP Labs.

12
 CLOUD COMPUTING

8. Advantages of Cloud Computing

The following are some of the possible advantages of cloud computing:

Flexibility

There is a high rate of flexibility when using cloud computing because people can opt out
of  using it whenever they want too. One of the  major benefit of cloud computing is
that there is no limitation of place and medium. We can access our applications and data
anywhere in the world, on any system. These are the main reasons, which attract people to
use  this method.

Low Cost

Companies can save big by employing cloud computing as it eliminates cost for hardware
and software. With cloud computing, company uses the resources of the hosting company to
store their data and applications.
Companies also pay for use of the software and programs by paying a subscription fee.

The cost of using cloud resources is very economical for resources such as centralized, real
estate, bandwidth, and power. Users will also save money on software updates,
management costs, and data storage costs. It is a cheaper way to maintain the software and
it will save time, as the developers keep track of updates and maintain your programs
while you use it.

Speed & Scales

Traditional methods to buy and configure hardware and software are time consuming. There
is no need to purchase and setup hardware manually when using the cloud computing
method. Cloud computing provides a rapid deployment model that enables applications to
grow quickly to match increasing usage requirements. Depending upon their needs the user
can quickly scale up or scale down.

Easier Management of Data and Information

Since all data are located on a centralized location, data are more organized making it easy
to manage. All transactions are also recorded so management can easily track activities of
their employees.

13
 CLOUD COMPUTING

Device Diversity

We can access our applications and data anywhere in the world, on any system. Cloud
Computing Services can be accessed through various different electronic devices that are
able to have access to the internet. These devices would include and i pad, Smartphone,
Laptop, or desktop computer.

Increased Storage Capacity

Increased Storage Capacity is another benefit of the cloud computing, as it can store more
data as compared to a personal computer. So it saves us from the upgrading computer
memory that helps reduce the cost for companies and users.

Easy to Learn and Understand

Thus allowing you to save hundreds and thousands of dollars in implementing any new
system and making arrangements for training on the same. Since people are quiet used to
cloud applications like Gmail, Google Docs, so anything related to the same is most likely
to be understood by the users.

Automatic Updating

It saves companies time and effort to update multiples server. On the other hand, it also
helps users to download updates for the software. Once the server gets updated the users
can get the updates without doing anything.

Customize Setting

Cloud computing also allows you to customize your business applications. This is a great
benefit because the world of online business is very competitive.

14
 CLOUD COMPUTING

9. Disadvantages of Cloud Computing

The following are some of the possible disadvantages of cloud computing:

Dependency

One major disadvantages of cloud computing is user’s dependency on the provider. Internet
users don’t have their data stored with them.

Risk

Cloud computing services means taking n services from remote servers.

A There is always insecurity regarding stored documents because users do not have
control over their software. Nothing can be recovered if their servers go out of service.

Requires a Constant internet connection

The most obvious disadvantage is that Cloud computing completely relies on network
connections.

It makes your business dependent on the reliability of your Internet connection. When it’s
offline, you’re offline. If you do not have an Internet connection, you can’t access anything,
even your own data. A dead internet connection means no work.

Similarly, a low-speed Internet connection, such as that found with dial-up services,
makes cloud computing painful at best and often impossible. Web- based apps often
require a lot of bandwidth to download, In other words, cloud computing isn’t for the slow
connection.

Security

Security and privacy are the biggest concerns about cloud computing companies might feel
uncomfortable knowing that their data is stored in virtual server which makes responsibility
on the security of the data difficult to determine and even users might feel uncomfortable
handing over their data to a third party

Privacy is another big issue with the cloud computing server. To make cloud servers more
secure to ensure that a clients data is not accessed by any unauthorized users, cloud service
providers have developed password protected accounts, security servers through which all

15
 CLOUD COMPUTING

data being transferred must pass and data encryption technique.

Migration Issue

Migration problem is also a big concern about cloud computing. If the user wants to
switch to some other Provider then it is not easy to transfer huge data from one provider to
another.

16
 CLOUD COMPUTING

10. Popular Cloud Applications

Applications using cloud computing are gaining popularity day by day for their high
availability, reliability and utility service model. Today many cloud providers are in the IT
market. Of those Google App-Engine, Windows Azure and Amazon EC2, S3 are prominent
ones for their popularity and technical perspective.

10.1 Amazon EC2 and S3 Services

Amazon Elastic Computing (EC2) is one of the biggest organizations to pro- vide
Infrastructure as a Service. They provide the computer architecture with XEN virtual
machine. Amazon EC2 is one of the biggest deployments of XEN architecture to date.
The clients can install their suitable operating system on the virtual machine. EC2 uses
Simple Storage Service (S3) for storage of data. Users can hire suitable amount CPU
power, storage, and memory without any upfront commitment. Users can control the
entire software stack from kernel upwards. The architecture has two components one
is the EC2 for computing purposes and S3 is for storage purposes

Simple
• Storage Service: S3 can be thought as a globally available distributed
hash table with high-level access control. Data is stored in name/value pairs.
Names are like UNIX file names and the value can be object having size up-to 5
GB with up-to 4K of metadata for each object. All objects in Amazon’s S3 must
fit into the global namespace. This namespace consists of a “bucket name” and an
“object name”. Bucket names are like user names in traditional email account and
provided by Amazon on first come first serve basis. An AWS (Amazon Web
Services) account can have maximum of 100 buckets.

Data to S3 can be sent by SOAP based API or with raw HTTP “PUT” com-
mands. Data can be retrieved using SOAP HTTP or Bit Torrent. While using Bit
Torrent the S3 system operates as both tracker and the initial seeder. There are
also some tools available which enables the users to view S3 as a remote file
system. Upload download rate from and to S3 is not that much exiting. One
developer from Germany reported experiencing 10-100 Kbps. This rate can go
17
 CLOUD COMPUTING

up-to 1-2 Mbps on the higher side depending on the time of the day. Although the
speed is not that much fascinating it is good enough for delivering web objects
and for backup purposes although for doing computation it is not suitable.

Amazon S3 has a very impressive support for privacy, integrity and short term
availability. Long term availability is unknown as this depends on the internal
commitment of Amazon data centers. Data privacy can be obtained by encrypting
the data to be stored. But this encryption is to be done by the user before storing
the data in S3. One can use SSL with HTTPS to connect to S3 for more security
but this usage of SSL increases upload/download time also. Data integrity can be
achieved by checking end to end MD5 checking. When an object is stored into S3
then it returns MD5 of that object. One can easily check it with previously
computed hash value to guarantee data integrity. Short term availability depends
upon the Amazon’s connectivity and load on its server at that instant. Once the
data is actually in the S3 then it is Amazon’s responsibility to take care of it’s
availability. They claim that the data is backed up on multiple hard drives in
multiple data centers but doesn’t guarantee this by any Service Level Agreement.
There is no backup or recovery mechanism if the user accidentally deletes any
data.

Amazon has a very impressive scheme of authentication in comparison to other

cloud services. Every AWS account has an Access Key ID and a Secret Key.

The ID is of 20 characters and the Key is a 41 character string. When signing

HMAC is first computed for the sign request parameters using that Key. And in
the Amazon server that HMAC is again computed and compared with the value
previously computed in the client side. These requests also include timestamp to
prevent replay attacks.

Elastic
• Compute Cloud: As the name implies EC2 rents cloud of computers to
the users with flexibility of choosing the configuration of the virtual machine like
RAM size, local disk size, processor speeds etc.

Machines that deliver EC2 services are actually virtual machines running on top
of XEN platform. Users can store a disk image inside S3 and create a virtual
18
 CLOUD COMPUTING

machine in EC2 using tools provided by Amazon. This virtual machine can be
easily instantiated using a java program and can also be monitored. As EC2 is
based on XEN it supports any linux distribution as well as other OSs. Amazon
does not promise about reliability of the EC2 computers. Any machine can crash
at any moment and they are not backed up. Although these machine generally
don’t crash according to the experience of the users but it is safe to use S3 to
store information which is more reliable and replicated service. EC2 security
model is similar to that of S3. The only difference is that the commands are
signed with an X 509 private key. But this key is downloaded from AWS account
so the security depends fundamentally on the AWS username and password.

10.2 Google App-Engine

Google App-Engine [1] is a platform for developing and deploying web applications in
Google’s architecture. This provides Platform as a Service to the cloud users. In 2008
Google App-Engine was first released as beta version. Languages supported by Google
App-Engine are python, java and any extension of JVM languages. App- Engine
requires developers to use only languages which are supported by it and this is also
applied with APIs and frameworks. Now Google App-Engine allows storing and
retrieving data from a BigTable non-relational database.

AppEngine applications are expected to be request-reply based. Google App-

engine provides automatic scalability, persistent data storage service. Data store features
a query engine and transaction capabilities. These applications are easy to scale as
traffic and data storage need to grow so the cloud user doesn’t have to worry about the
spikes in the traffic or data. These applications are generally suitable for social
networking start-ups, event-based websites catering to seasonal events or institutions
(schools, colleges, universities, government agencies) etc.

10.3 Windows Azure

Windows Azure is an intermediate in the spectrum of flexibility vs programmer

convenience. These systems use .NET libraries to facilitate language independent
19
 CLOUD COMPUTING

managed environment. This service falls under the category of Platform as a Ser- vice.
Though it is actually in between complete application framework like Google App-
Engine and hardware virtual machines like EC2. Azure applications run on machines in
Microsoft data centers. By using this service customers can use it to run applications
and store data on internet accessible machines owned by Microsoft. windows Azure
platform provides three fundamental components - compute component, storage
component and fabric component. Basic components of Windows Azure are shown in
Figure.

The
• Compute Service: The primary goal of this platform is to support a large
number of simultaneous users. (Microsoft also said that they would use Azure to
build their SaaS applications which motivated many potential users.) To allow
applications to scale out Microsoft uses multiple instances of that appli- cations
on virtual machines provided by Hypervisor. Developers use Windows Azure
portal through Web browser, and use Windows live ID to sign in into his/her
hosting account or storage account or both.

Two different types of Azure instance is available: Web role instance and Worker
role instances.

Web role instance: As the name implies this type of instance can accept
HTTP or HTTPS requests. For this facility Microsoft uses IIS (Internet
Information Services) as a web server inside the VM provided. Developers
can build applications using ASP.NET, Windows Communication Foun-
dation (WCF) or any other .NET technology or native codes also like C++.
PHP or java based technologies also supported in Azure. Azure scales
applications by running multiple instances without any affinity with a
particular Web role instance. So it is perfectly natural for an Azure
application to serve multiple requests from a single user by multiple in-
stances. So this requires to write the client state in the Azure storage after
each client request.

20
 CLOUD COMPUTING

Worker role instance: These types of instances are very similar to that of
Web role instances. But unlike the Web role instances these don’t have

21
 CLOUD COMPUTING

Windows Azure component architecture

IIS configured. They can be configured to run executable of users’ right. Worker role
instance is more likely to function like a background job. Web role instances can be used
to accept request from the users and then they can be processed by Worker role instances
in a later point of time. For a compute intensive work many Worker role instances can run
in parallel.

Logging and monitoring of Azure applications is made easy by provision of ap- plication
wide log. a developer can collect performance related information like measure of CPU
usage, store crash dumps in the storage. Azure doesn’t give the developer the freedom to
use his/her own VM image for Windows Azure. The platform maintains its own Windows.
Applications in Azure run only in user mode - no administrative access isn’t allowed here.
So Windows Azure can update the operating system in each VM without any concern of
affecting the applications running on it. This approach separates administrative work from
the user domain.

22
 CLOUD COMPUTING

The Storage Service: Applications running in Azure uses storage of different types:
•
Blobs: This is used for storing binary data in a simple hierarchy. Blobs can
have associated metadata with them. A user account can have one or more
containers and these containers have one or more blobs.

Storage tables: Blobs provide mechanisms for unstructured data but for
more structured purposes tables are more suitable. These tables are nothing
like tables in a traditional database. They are actually stored in a group of
entities. These tables can be accessed by using ADO.NET Data Services.
SQL is not preferable for scale out issues.

Queue : This is not a structure like tables or blobs to store data but these
queues are used to store messages about tasks to be performed by Worker
role instance. These tasks are written by Web role instances on receiving
request from clients. A Worker role instance waiting on that queue can read
the message and perform the task it specifies.
All data in the Windows Azure storage is replicated three times for providing fault
tolerance. Azure also keeps backups in geographically distributed data centers. Windows
Azure storage can be accessed by any Windows Azure ap- plication as well as any
application hosted at another cloud platform. All the blobs, tables, queues are named using
URIs and can be accessed by HTTP methods calls.

Some applications have inherent need for relational databases. This is provided in the form
of SQL Azure. This is build on Microsoft SQL Server. This data can be accessed via
ADO.NET or by other Windows data access interfaces.

The Fabric:
• All Windows Azure application and all of the data stored in Azure Storage
live are physically happening inside some of the data centers handled by Microsoft. In the
data centers the set of machines dedicated to Azure are organized into a fabric. These
machines are managed by fabric controller. These are replicated in five to seven machines.
These controllers are aware of every Windows Azure application running in that fabric and
also own all the resources like computers, switches, load balancers etc. Controllers
monitors, decides which resources to allocate to new applications looking at the

23
 CLOUD COMPUTING

configuration file with the application. They also monitor the running applications.

24
 CLOUD COMPUTING

11. Conclusion

So, while cloud computing is really great and you’re probably already using it, either for
business of for personal means, here’s what we’ve learned from taking a look at the pros
and cons:

 Cloud computing is a really cheap way for companies to have all the resources
they need in once place.
 It’s a much better way to spread your resources, and it becomes easier to access
things from longer distances.

But…

 Is cloud computing really all that safe? For instance, while cloud computing has
backups, if something was to go wrong… do you have your own personal backup?
 Is cloud computing as private as you think it is? Is your data secured or can
anyone access it?

So there you have it! Everything you need to know about cloud computing; what it is, how
to use it, and some pros and cons!

25
 CLOUD COMPUTING

12. References
Magazines
Cloud Computing – MLADEN .A.VOUK -Issues, Research an Implementations,
Information Technical Interfaces, June 2008
Websites:
www.google.com www.wilkipedia.com

www.ibm.com/developerworks/websphere/zones/hipods/library.html\

http://communication.howstuffworks.com/cloud-computing2.html

www.thbs.com

http://www.wikipedia.org/

26