INTRODUCTION

SUN Microsystems`s slogan was 'The network is the computer and that`s as good as any to
describe how Cloud Computing works. In this report it is understood that how Cloud Computing
helps us to use the network as computer.
From Past to Aow
Many years ago when the electricity revolution occurred, it was so expensive and everyone who
wanted to use this resources, used Ireestanding generators to produce electricity power. AIter
that people and businesses shutdown their generators and bought electricity Irom utilities.
Nowadays we experience new revolution, Cloud Computing. Such as electricity, Cloud
Computing invites us to change the way that we think about computing: user-centric instead oI
desktop-centric. It means that the potential oI the Cloud is used to transIer the computing and
processing Irom locality into the Cloud, so the capacity oI the Cloud is used in the way that we
didn`t Iace beIore.
ow Cloud is made?
The key to Cloud Computing is the 'Cloud, a massive network oI servers or even individual
PCs interconnected in a grid (as in Fig. 1). In Iact the most proliIic Cloud which can be used is
the Internet. So in this article Cloud is a metaphor Ior the Internet. The important element in
Cloud is that human management isn't needed Ior allocating processes to resources. As you see
in the Fig. 1, since hardware is in the Cloud, it`s invisible. This Cloud is simple; however it
needs to be managed Ior working intelligently to approach the task processing.

Figure 1. How users connect to the cloud

C efinition of Cloud Computing
Cloud Computing is a paradigm that Iocuses on sharing data and computations over a scalable
network oI nodes. Examples oI such nodes include end user computers, data centers, and Cloud
Services. We term such a network oI nodes as a Cloud. (as in Fig. 1). Cloud Computing is the
next generation in computing. It is a type oI computing that is comparable to grid Computing,
relies on sharing computing resources rather than having local servers or personal devices to
handle applications. The goal oI Cloud Computing is to apply traditional super computing power
(normally used by military and research Iacilities) to perIorm tens oI trillions oI computations
per second. The memorable point about this advantage is that this perIormance can be produced
by a grid oI computing resources that work together or even by the Iorce oI the low-cost PCs,
like Personal PCs. In Iact you get super-computing power at PC prices. It`s similar to the rain
power that is made by water drops and the Cloud has signiIicant role in this event! The main idea
is to use the existing inIrastructure in order to bring all Ieasible services to the Cloud and make it
possible to access those services regardless oI time and location.
ainers
The main idea in Cloud Computing is that it covers all the range oI users, Irom home users that
use Cloud Computing to approve their works better and IT staIIs and enterprise managers that
use Cloud Computing Ior optimizing, planning and implementing their enterprises. So we can
Iace it that Cloud Computing has the big roles in our liIe in the Iuture.

T IS CLOUD COMPUTING ?

et`s Iirstly start by establishing what the word 'cloud means in this context.

Cloud is a metaphor Ior the Internet. It is quite common these days to draw network diagrams
that depict the Internet as a cloud hence the use oI the word in this instance. So why didn`t they
just call it 'Internet Computing? Well that`s because cloud computing goes beyond just using
the Internet, as we do Ior, say, websites. The Internet is the carrier but there is a whole set oI
other technologies involved that are essential in a cloud computing environment. Getting a
commonly agreed deIinition oI cloud computing is still quite diIIicult. There are many opinions
on this, some oI them driven by vendors` own view oI the world. Whilst they all generally reIer
to the same thing we thought we should try and agree on one oI these Ior the purposes oI this
guide. We like the Gartner deIinition which deIines cloud computing as

a style of computing where massively scalable I1-enabled capabilities are delivered :as a
service' to external customers using Internet technologies

The three key words in this deIinition are scalable, service and Internet. Cloud computing is
about how an application is deployed and delivered over the Internet and which is scalable on
demand.
Cloud computing is not something that an end-user buys. In Iact, end-users should be oblivious
to, and shouldn`t care, whether an application is delivered using cloud computing. Cloud
computing is a deployment model Ior applications that is used by organisations in order to
reduce inIrastructure costs and/or address capacity/scalability concerns. EIIectively these
organisations are saying that they don`t want to own the assets or to operate the system in their
own data centres. They are buying results not assets (oI course, it is still possible to own the
intellectual property in the soItware and to have it operating in the cloud).

Cloud computing has many diIIerent meanings Ior diIIerent people. However, a basic deIinition
that encompasses virtually all deIinitions is the Iollowing: Cloud computing is a modality oI
computing characterized by on demand availability oI resources in a dynamic and scalable
Iashion. The term resource here could be used to represent inIrastructure, platIorms, soItware,
services, or storage. The cloud provider is responsible to make the needed resources available on
demand to the cloud users. It is the responsibility oI the cloud provider to manage its resources in
an eIIicient way so that the user needs can be met when needed at the desired Quality oI Service
(QoS) level. For example, an inIrastructure cloud oIIers computing inIrastructure, typically in the
Iorm oI virtual machines allocated to physical servers, as needed by its users. Cloud users are
charged, in most cases, by how much resources they consume (e.g., $ per hour oI CPU usage).
An analogy between cloud computing and the power grid is useIul to capture some oI the
similarities but also some important distinctions. Consumers use electric energy on-demand
according to their needs and pay based on their consumption. The electric power utilities
(analogous to cloud providers) have to be able to determine at each point in time how much
energy to generate in order to be able to match the demand. A variety oI models that include
weather prediction models and historic data on power consumption Ior each day oI the year and
each hour oI the day drive the decisions made by utilities. The analogy breaks in some oI
important aspects. The product delivered by the power grid is homogeneous (e.g., 110 V oI
alternating current at 60 Hz). On the other hand, computing clouds oIIer a variety oI resources on
demand. Another important diIIerence has to do with the interIace and plug-compatibility. One
can plug any appliance to the power grid and it will work seamlessly as long as it conIorms to a
very simple speciIication oI voltage and Irequency. The same is not true with computing clouds.
The APIs oIIered by cloud providers are not standardized and may be very complicated in many
cases. In other words, computing clouds are not 'plug-and-play yet.
The Iollowing are some oI the advantages oI cloud computing:
1. !ay as you go:
Companies can avoid capital expenditures by using cloud resources on an as needed basis. In the
owned approach, the total cost oI ownershipm (TCO) is a Iunction oI three main components:
O initial capital investment Ior hardware, soItware, networking, Iacilities inIrastructure
including cooling and power,
O operational costs which includes hardware and soItware maintenance, personnel cost
(including system, network, database administrators, capacity planning analysts), power
consumption, and depreciation,
O system upgrades required to cope with the growth oI existing workload and/or new
workloads.

2. o need to provision for peak loads:
II cloud computing resources are used, the responsibility to support peak loads at agreed upon
service levels rests with the cloud computing provider.
3. %ime to market :
Because users oI cloud computing resources do not need to procure, install, and test all the
inIrastructure including middleware and applications in many cases, they can be up and running
in very little time.
4. onsistent performance and availability:
When services are provided by the cloud under strict SAs that are speciIic on response time and
availability, users do not need to worry so much about maintaining adequate levels Ior these
metrics. This burden is shiIted to the cloud, which, by virtue oI managing a typically large
inIrastructure may be able to autonomically shiIt resources (e.g., virtual machines) to keep up
with varying and unpredictable workloads.

The potential drawbacks and or concerns regarding cloud computing are:
1. !rivacy and security:
Many organizations may be concerned about having their sensitive data living in the same
platIorms as that oI their competitors. There may be concerns regarding exposing a company`s
private data to the cloud computing provider. In some cases, a company may be bound to several
types oI regulations (e.g., HIPPA) whose responsibility cannot be easily delegated to a third
party provider.

2. ternal dependency for mission critical applications:
Even when cloud providers oIIer to adhere to strict SAs and pay penalties Ior non compliance,
cloud users may be concerned about trusting some oI their mission critical applications to a third
party.
3. isaster recovery:
Users oI resources in a cloud need to have guarantees that the provider has adequate backup and
disaster recovery plans that will prevent a disruption oI a user`s activities in the Iace oI natural or
man-made disasters.
4. onitoring and nforcement of SLAs:
Negotiating, monitoring, and enIorcing SAs may be challenging in cloud computing because
cloud resources and services are shared by a multitude oI users and because providers have little
control over the workload intensity oI the diIIerent cloud applications.
Cloud computing provides computation, soItware, data access, and storage services that do not
require end-user knowledge oI the physical location and conIiguration oI the system that delivers
the services. Parallels to this concept can be drawn with the electricity grid, wherein end-users
consume power without needing to understand the component devices or inIrastructure required
to provide the service. The concept oI cloud computing Iills a perpetual need oI IT: a way to
increase capacity or add capabilities on the Ily without investing in new inIrastructure, training
new personnel, or licensing new soItware. Cloud computing encompasses any subscription-based
or pay-per-use service that, in real time over the Internet, extends IT's existing capabilities.
Cloud computing describes a new supplement, consumption, and delivery model Ior IT services
based on Internet protocols, and it typically involves provisioning oI dynamically scalable and
oIten virtualized resources. It is a byproduct and consequence oI the ease-oI-access to remote
computing sites provided by the Internet. This may take the Iorm oI web-based tools or
applications that users can access and use through a web browser as iI they were programs
installed locally on their own computers. Cloud computing providers deliver applications via the
internet, which are accessed Irom a Web browser, while the business soItware and data are
stored on servers at a remote location. In some cases, legacy applications (line oI business
applications that until now have been prevalent in thin client Windows computing) are delivered
via a screen-sharing technology, while the computing resources are consolidated at a remote data
center location; in other cases, entire business applications have been coded using web-based
technologies such as AJAX. Most cloud computing inIrastructures consist oI services delivered
through shared data-centers and appearing as a single point oI access Ior consumers' computing
needs. Commercial oIIerings may be required to meet service level agreements (SAs), but
speciIic terms are less oIten negotiated by smaller companies.


CLOUD COMPUTING LOGICL DIGRM


CLOUD COMPUTING RCITECTURE

Several Iundamental components make up the cloud architecture:

O Computing resources are located oII site in a data center that is not owned or
managed by the enterprise using the cloud services.
O #esources oIten leverage virtualization Ior ease oI management and
interoperability.
O #esources are available on demand.
O InIrastructure is oIten shared.
O Virtualization can enable multiple customers and applications to share the same
physical machines.
O Services are generally provisioned on demand and scaled up or down as required.
O Services are usually subscription-based, with a variety oI tiered service oIIerings
as well as Ilat-rate and per-use pricing models.

These components, Iundamentally tied together into an architecture, produce a cloud services
oIIering. The architecture oI cloud computing can be described using a layered model, in a
manner similar to that oI the Open Systems Interconnection (OSI) seven-layer model developed
to provide an abstract description oI layered communications and computer network protocol
design. At the top oI the cloud model is the client layer, which interIaces directly with cloud
environment end users. Below the client layer is the cloud applications layer. Applications that
run on the cloud reside here and are generally accessed by application developers. Next is the
soItware inIrastructure layer, where basic inIrastructure services, including storage, computing,
and communications, are perIormed. Below these three layers are the actual cloud environment
soItware and hardware layers. At the soItware layer resides the kernel that translates and
executes the cloud applications` instructions on the cloud hardware. In many architectures, this
cloud soItware kernel can include a hypervisor Ior executing virtualized applications. Finally,
underpinning all oI the cloud layers is the hardware layer, which includes processor, memory,
storage, and communications hardware. Figure 2 depicts the relevant layers.



Figure 2. Cloud ayers Model

As you can see in Figure 3., at Iirst, user select a Cloud service under the InterIace, then the
System Management choose the appropriate service, next this service is started and at last
launches appropriate data and web application.















Figure 3. The architecture behind Cloud Computing





It was just a simple sample oI Cloud Computing architecture to show a way how it works; Many
researches are doing now to make Cloud Computing more intelligent by means oI utonomic
Computing to make this architecture more eIIicient.

When talking about a cloud computing system, it's helpIul to divide it into two sections: the Iront
end and the back end. They connect to each other through a network, usually the Internet. The
Iront end is the side the computer user, or client, sees. The back end is the "cloud" section oI the
system.
The Iront end includes the client's computer (or computer network) and the application required
to access the cloud computing system. Not all cloud computing systems have the same user
interIace. Services like Web-basede-mail programs leverage existing Web browsers like Internet
Explorer or FireIox. Other systems have unique applications that provide network access to
clients.

On the back end oI the system are the various computers, servers and data storage systems that
create the "cloud" oI computing services. In theory, a cloud computing system could include
practically any computer program you can imagine, Irom data processing to video games.
Usually, each application will have its own dedicated server.
A central server administers the system, monitoring traIIic and client demands to ensure
everything runs smoothly. It Iollows a set oI rules called protocols and uses a special kind oI
soItware called middleware. Middleware allows networked computers to communicate with each
other.
II a cloud computing company has a lot oI clients, there's likely to be a high demand Ior a lot oI
storage space. Some companies require hundreds oI digital storage devices. Cloud computing
systems need at least twice the number oI storage devices it requires to keep all its clients'
inIormation stored. That's because these devices, like all computers, occasionally break down. A
cloud computing system must make a copy oI all its clients' inIormation and store it on other
devices. The copies enable the central server to access backup machines to retrieve data that
otherwise would be unreachable. Making copies oI data as a backup is called redundancy.











COMPRISON OF CLOUD COMPUTING IT :

O Autonomic computing Computer systems capable oI selI-management."
O Clientserver model lientServer omputing reIers broadly to any distributed
application that distinguishes between service providers (servers) and service requesters
(clients).
O Grid computing "A Iorm oI distributed computing and parallel computing, whereby a
'super and virtual computer' is composed oI a cluster oI networked, loosely coupled
computers acting in concert to perIorm very large tasks."
O MainIrame computer PowerIul computers used mainly by large organizations Ior
critical applications, typically bulk data processing such as census, industry and
consumer statistics, enterprise resource planning, and Iinancial transaction processing.
O Utility computing The "packaging oI computing resources, such as computation and
storage, as a metered service similar to a traditional public utility, such as electricity."
O Peer-to-peer Distributed architecture without the need Ior central coordination, with
participants being at the same time both suppliers and consumers oI resources (in contrast
to the traditional clientserver model).
O Service-oriented computing Cloud computing provides services related to computing
while, in a reciprocal manner, service-oriented computing consists oI the computing
techniques that operate on soItware-as-a-service.

CLOUD OFFERING

Cloud computing is really about two Iundamental concepts: leveraging economies oI scale and
improving hardware use. In its current buzzword status, many cloud oIIerings are quick to
associate a service or solution with the cloud architecture. While taking many Iorms, cloud
oIIerings can generally be sorted into several categories, including:

Software as a Service
SaaS allows an application to be delivered through a Web browser. The application is generally
soItware and hardware agnostic and relies on server components outside the user`s network. The
server hardware can be owned and managed by the Iirm selling the SaaS application, or it can be
Iurther removed (in other words, hosted and managed by another party). SaaS is generally
characterized in Iour maturity levels, with the most mature allowing the greatest Ilexibility,
scalability, and reliability. Traditional soItware Iirms are beginning to provide SaaS oIIerings to
small- and medium-sized enterprises that oIten do not have the inIrastructure and resources to
run larger-scale enterprise applications. Other SaaS oIIerings include typical productivity and
oIIice suites, which can reduce an enterprise`s licensing, maintenance, and inIrastructure
requirements.
Utility Computing
The term utility computing predates widespread use oI the term cloud, but is a classic example oI
the beneIits oI a cloud inIrastructure. Providers oIIer solutions that enable a virtual data center
that is, an Internet-enabled commodity processing and storage hardware environment.
Enterprises can outsource some or all data center needs to the utility computing provider, oIten
starting with lower-value, routine processing and storage. Current oIIerings range Irom on-
demand processing and storage up to entire remotely hosted and managed data centers. The
cloud paradigm allows these services to be oIIered via a distributed, Ilexible, connected
topology, rather than through a single data center.
isaster Recovery
Along with oIIering traditional server replacement and utility computing options, the cloud also
enables a new way to deliver disaster recovery services. Disaster recovery oIten requires
dedicated, speciIic hardware Ior data storage and remote applications operation. The cloud
paradigm can allow enterprises to to replace dedicated data centers Ior disaster recovery, thereby
reducing the costs to provide this service and, potentially, making it available to a wider range oI
clients.
pplication Programming Interfaces
APIs are becoming a popular way to provide new service oIIerings while leveraging the cloud
inIrastructure. APIs allow unique applications to be written and oIIered via the Web using
existing soItware and services. These applications are soItware and hardware agnostic and can be
run via a Web browser. An application provider can use another provider`s Web-enabled
services and soItware and either manage and host the application itselI, or allow the existing
provider to host the application on its own network.
Managed Services
Managed services, such as remote monitoring and administration, security services, anti-virus
scanning, and other back-end oIIerings, have been prominent Ior over a decade. However, recent
advances allow services oIIerors to distribute the deployment and management oI their oIIerings,
resulting in Ilexibility, increased reliability, and reduced costs Ior both the provider and the user.

Large ataset Processing
Cloud architectures have enabled signiIicantly improved processing oI large datasets. The
commercially developed Google Map#educea programming model and associated
implementation Ior processing and generating large datasetsis particularly eIIicient and allows
distribution, independent processing, and consolidation oI analysis, as shown in. Apache
Hadoop is a soItware product that provides a distributed computing platIorm Ior sharing and
processing large amounts oI data. The Hadoop project develops open-source soItware Ior
reliable, scalable, distributed computing; one oI its subprojects is a Iree, open-source version oI
Map#educe, available Ior general use. In a recent example oI its application, an enterprise
needed to process terabytes oI data and analyze server use logs to provide better troubleshooting
and optimization. When the datasets became too large Ior a single, high-perIormance machine to
handle eIIiciently, the enterprise used Hadoop (including Map#educe) and other Apache
products to deploy and process the data across 10 commodity nodes. PerIormance was greatly
improved, processing times were reduced, and the system demonstrated signiIicant scalability.

Cloud Integration
While a variety oI cloud services scenarios and deployments exist, many cloud oIIerings are still
deployed as 'islands today. Enterprises may have to integrate data and existing applications
with applications deployed on a cloud or integrate multiple cloud applications. A new cloud
integration market is emerging, with established and upstart players oIIering soItware and
services that promise to stitch together disparate applications and services in a cloud
environment.






Table 1 summarizes the beneIits to be derived
Irom taking advantage oI the seven cloud
oIIerings just discussed.



CRCTERISTICS OF CLOUD COMPUTING

Cloud computing exhibits the Iollowing key characteristics:

O gility improves with users' ability to re-provision technological inIrastructure resources.
O pplication Programming Interface (API) accessibility to soItware that enables
machines to interact with cloud soItware in the same way the user interIace Iacilitates
interaction between humans and computers. Cloud computing systems typically use
#EST-based APIs.
O Cost is claimed to be reduced and in a public cloud delivery model capital expenditure is
converted to operational expenditure. This is purported to lower barriers to entry, as
inIrastructure is typically provided by a third-party and does not need to be purchased Ior
one-time or inIrequent intensive computing tasks. Pricing on a utility computing basis is
Iine-grained with usage-based options and Iewer IT skills are required Ior
implementation (in-house).
O evice and location independence enable users to access systems using a web browser
regardless oI their location or what device they are using (e.g., PC, mobile phone). As
inIrastructure is oII-site (typically provided by a third-party) and accessed via the Internet,
users can connect Irom anywhere.
O Multi-tenancy enables sharing oI resources and costs across a large pool oI users thus
allowing Ior:
Centralization oI inIrastructure in locations with lower costs (such as real estate,
electricity, etc.)
Peak-load capacity increases (users need not engineer Ior highest possible load-
levels)
Utilization and efficiency improvements Ior systems that are oIten only 1020
utilized.
O Reliability is improved iI multiple redundant sites are used, which makes well-designed
cloud computing suitable Ior business continuity and disaster recovery.
O Scalability and Elasticity via dynamic ("on-demand") provisioning oI resources on a
Iine-grained, selI-service basis near real-time, without users having to engineer Ior peak
loads.
O Performance is monitored, and consistent and loosely coupled architectures are
constructed using web services as the system interIace.
O Security could improve due to centralization oI data, increased security-Iocused
resources, etc., but concerns can persist about loss oI control over certain sensitive data,
and the lack oI security Ior stored kernels. Security is oIten as good as or better than
under traditional systems, in part because providers are able to devote resources to
solving security issues that many customers cannot aIIord. However, the complexity oI
security is greatly increased when data is distributed over a wider area or greater number
oI devices and in multi-tenant systems that are being shared by unrelated users. In
addition, user access to security audit logs may be diIIicult or impossible. Private cloud
installations are in part motivated by users' desire to retain control over the inIrastructure
and avoid losing control oI inIormation security.
O Maintenance oI cloud computing applications is easier, because they do not need to be
installed on each user's computer. They are easier to support and to improve, as the
changes reach the clients instantly.

CLLENGES TO BE FCED
O Complete Security
O Good PerIormance
O Available Everywhere
O It Is Hard to Integrate With
O In-House IT Still Not Possible
O Not Enough Ability to Customize
O Worried That Cloud Will Cost More
O Bringing Back In House May Be DiIIicult
O Not Enough Major Suppliers Yet
CLOUD STORGE

Cloud storage is a model oI networked online storage where data is stored on multiple virtual
servers, generally hosted by third parties, rather than being hosted on dedicated servers. Hosting
companies operate large data centers; and people who require their data to be hosted buy or lease
storage capacity Irom them and use it Ior their storage needs. The data center operators, in the
background, virtualize the resources according to the requirements oI the customer and expose
them as storage pools, which the customers can themselves use to store Iiles or data objects.
Physically, the resource may span across multiple servers. Cloud storage services may be
accessed through a web service application programming interIace (API), or through a Web-
based user interIace.

Cloud storage advantages
O Companies need only pay Ior the storage they actually use.
O Companies do not need to install physical storage devices in their own datacenter or
oIIices, which reduces IT and hosting costs.
O Storage maintenance tasks, such as backup, data replication, and purchasing additional
storage devices are oIIloaded to the responsibility oI a service provider, allowing
organizations to Iocus on their core business.

Potential concerns
O Security oI stored data and data in transit may be a concern when storing sensitive data
at a cloud storage provider
O PerIormance may be lower than local storage
O #eliability and availability depends on wide area network availability and on the level oI
precautions taken by the service provider.
O Users with speciIic records-keeping requirements, such as public agencies that must
retain electronic records according to statute, may encounter complications with using
cloud computing and storage.
LERS IN CLOUD COMPUTING



Once an Internet Protocol connection is established among several computers, it is possible
to share services within any one oI the Iollowing layers:
O Client
A cloud client consists oI computer hardware and/or computer soItware that relies on
cloud computing Ior application delivery and that is in essence useless without it.
Examples include some computers, phones and other devices, operating systems, and
browsers.

O pplication
Cloud application services or "SoItware as a Service (SaaS)" deliver soItware as a
service over the Internet, eliminating the need to install and run the application on the
customer's own computers and simpliIying maintenance and support.

O Platform
Cloud platIorm services, also known as PlatIorm as a Service (PaaS), deliver a
computing platIorm and/or solution stack as a service, oIten consuming cloud
inIrastructure and sustaining cloud applications. It Iacilitates deployment oI applications
without the cost and complexity oI buying and managing the underlying hardware and
soItware layers.

O Infrastructure
Cloud inIrastructure services, also known as InIrastructure as a Service (IaaS), deliver
computer inIrastructure typically a platIorm virtualization environment as a service,
along with raw (block) storage and networking. #ather than purchasing servers, soItware,
data-center space or network equipment, clients instead buy those resources as a Iully
outsourced service. Suppliers typically bill such services on a utility computing basis;
the amount oI resources consumed (and thereIore the cost) will typically reIlect the level
oI activity

O Server
The servers layer consists oI computer hardware and/or computer soItware products that
are speciIically designed Ior the delivery oI cloud services, including multi-core
processors, cloud-speciIic operating systems and combined oIIerings.

Figure 4. ayers in cloud computing

CLOUD SERVICES (CTEGORIES)

O Software as a Service (SaaS)

Software as a service (SaaS) , sometimes reIerred to as "on-demand soItware," is a
soItware delivery model in which soItware and its associated data are hosted centrally
(typically in the (Internet cloud) and are typically accessed by users using a thin client,
normally using a web browser over the Internet.
SaaS has become a common delivery model Ior most business applications, including
accounting, collaboration, customer relationship management (C#M), enterprise resource
planning (E#P), invoicing, human resource management (H#M), content management
(CM) and service desk management. SaaS has been incorporated into the strategy oI all
leading enterprise soItware companies.
According to a Gartner Group estimate, SaaS sales in 2010 have reached $10B, and are
projected to increase to $12.1b in 2011, up 20.7 Irom 2010. Gartner Group estimates
that SaaS revenue will be more than double its 2010 numbers by 2015 and reach a
projected $21.3b. Customer relationship management (C#M) continues to be the largest
market Ior SaaS. SaaS revenue within the C#M market is Iorecast to reach $3.8b in 2011,
up Irom $3.2b in 2010.
O Infrastructure as a Service
Infrastructure as a Service (IaaS) describes one oI the three main methods oI
accessing cloud computing based services. Organisations rent computing power and disk
space and access them Irom desktop PCs through a private network or across the internet.
IaaS has its roots in thin computing techniques that have been developing since the mid
1990s. But it was not adopted widely until virtualisation and Iast, reliable and widespread
internet access made it workable. Cloud computing in general reIers to the idea that a
company can rent space Irom a service provider and use it across the internet rather than
having an IT inIrastructure on their own premises.
With IaaS the enterprise renting the service is likely to have considerable IT skills in-
house because the inIrastructure oIIered is quite plain. Customers select and basic
soItware servers Ior their part oI the cloud and then load up their libraries, applications
and data then conIigure them themselves. Virtualisation enables IaaS providers to oIIer
almost unlimited instances oI servers to customers and make cost-eIIective use oI the
hosting hardware.
Organisations can use IaaS to quickly build new versions oI applications or environments
without having to order new hardware then wait Ior it to arrive and be conIigured.
Another popular use oI IaaS is hosting the websites oI organisations. This keeps the
website and its drain on IT resources away Irom an internal inIrastructure whose primary
purpose is to run the business, not the website. In these instances the IaaS provider takes
on any worries about monitoring traIIic and keeping the website available.
ike most cloud computing service oIIerings IaaS is priced on a pay-as-you-go model,
with organisations paying Ior the services provided Irom a menu. There is usually a price
Ior the number oI 'instances' oI servers, either Windows or Unix, and the soItware
environments loaded onto them such as databases or web servers, application
development environments or others. Then charges are made Ior the amounts oI data
hosted and network traIIic in and out oI the servers.
Amazon Web Services (AWS) is the most successIul IaaS provider to date. Amazon has
huge server Iarms that support the Amazon website and plenty oI experience with
running a reliable multiple server operation. It was decided to take the skills, technologies
and hardware they had developed or acquired and oIIer them Ior hire to other
organisations.


O Platform as a Service
Platform as a service (PaaS) is the delivery oI a computing platIorm and solution stack
as a service. PaaS oIIerings Iacilitate deployment oI applications without the cost and
complexity oI buying and managing the underlying hardware and soItware and
provisioning hosting capabilities, providing all oI the Iacilities required to support the
complete liIe cycle oI building and delivering web applications and services entirely
available Irom the Internet. PaaS oIIerings may include Iacilities Ior application design,
application development, testing, deployment and hosting as well as application services
such as team collaboration, web service integration and marshalling, database integration,
security, scalability, storage, persistence, state management, application versioning,
application instrumentation and developer community Iacilitation. These services may be
provisioned as an integrated solution over the web.
Types
O Add-on development Iacilities : These Iacilities allow customization oI existing
soItware-as-a-service (SaaS) applications, and in some ways are the equivalent oI
macro language customization Iacilities provided with packaged soItware
applications such as otus Notes, or MicrosoIt Word. OIten these require PaaS
developers and their users to purchase subscriptions to the co-resident SaaS
application.
O Stand alone development environments : Stand-alone PaaS environments do not
include technical, licensing or Iinancial dependencies on speciIic SaaS
applications or web services, and are intended to provide a generalized
development environment.
O Application delivery-only environments : Some PaaS oIIerings lack development,
debugging and test capabilities, and provide only hosting-level services such as
security and on-demand scalability.
O Open platIorm as a service : ets the developer use any programming language,
any database, any operating system, any server, etc.

TPES OF CLOUD

Cloud Computing can be classiIied into 4 types on the basis oI location where the cloud is
hosted:
O Public Cloud: Computing inIrastructure is hosted at the vendor`s premises. The customer
has no visibility over the location oI the cloud computing inIrastructure. The computing
inIrastructure is shared between organizations.
O Private Cloud: Computing architecture is dedicated to the customer and is not shared
with other organisations. They are expensive and are considered more secure than Public
Clouds. Private clouds may be externally hosted ones as well as in premise hosted clouds.
diagram is available under a Creative Commons Attribution ShareAlike 3.0 license in
PNG and SVG Iormats Irom the Wikimedia Commons (Cloud computing types.svg




Figure 5. Collaborations oI diIIerent clouds.
O ybrid Cloud: Organisations host some critical, secure applications in private clouds.
The not so critical applications are hosted in the public cloud. The combination is known
as Hybrid Cloud. Cloud bursting is the term used to deIine a system where the
organisation uses its own inIrastructure Ior normal usage, but cloud is used Ior peak
loads.
O Community Cloud: The cloud inIrastructure is shared between the organizations oI the
same community. For example, all the government agencies in a city can share the same
cloud but not the non government agencies.
CONCLUSION

The potential pitIalls associated with cloud service deployment have given pause to some major
cloud services providers. As a result, several oI these providers, along with enterprises wishing
to beneIit Irom cloud computing, have joined together to publish the Open Cloud ManiIesto.
This document is intended to encourage a dialogue among the providers and users oI cloud
computing services about the inIrastructure requirements and the need Ior open, interoperable
standards, including appropriate existing and adopted standards, as well as new standards when
warranted.

Building a trust over cloud computing will help in the development oI this technology. The
Iuture oI computing , need satisIactory trust, so that it could be implemented and enhanced
properly.










REFERENCES

O "Gartner Says Cloud Computing Will Be As InIluential As E-business" (http:/ / www.
gartner. com/ it/ page. jsp?id707508). Gartner.com. . #etrieved 2010-08-22.
O Gruman, Galen (2008-04-07). "What cloud computing really means" (http:/ / www.
inIoworld. com/ d/ cloud-computing/ what-cloud-computing-really-means-031).
InfoWorld. . #etrieved 2009-06-02.
O "Cloud Computing: Clash oI the clouds" (http:/ / www. economist. com/ displaystory.
cIm?storyid14637206). The Economist. 2009-10-15. #etrieved 2009-11-03.
O Cloud Computing DeIined (http:/ / www. cloudcomputingdeIined. com/ ) 17 July 2010.
#etrieved 26 July 2010.
O Buyya, #ajkumar; Chee Shin Yeo, Srikumar Venugopal (PDF). arket-Oriented loud
omputing. Jision, Hype, and Reality for elivering I% Services as omputing Utilities
(http:/ / www. gridBus. org/ ~raj/ papers/ hpcc2008keynotecloudcomputing. pdI).
Department oI Computer Science and SoItware Engineering, University oI Melbourne,
Australia. p. 9. #etrieved 2008-07-31.
O "What's In A Name? Utility vs. Cloud vs Grid" (http:/ / www. datacenterknowledge.
com/ archives/ 2008/ Mar/ 25/ whatsinanameutilityvscloudvsgrid. html).
Datacenterknowledge.com. . #etrieved 2010-08-22.
O "Distributed Application Architecture" (http:/ / java. sun. com/ developer/ Books/ jdbc/
ch07. pdI). Sun Microsystem. . #etrieved 2009-06-16.
O "Sun CTO: Cloud computing is like the mainIrame" (http:/ / itknowledgeexchange.
techtarget. com/ mainIrame-blog/ sun-cto-cloud-computing-is-like-the-mainIrame/ ).
Itknowledgeexchange.techtarget.com. 2009-03-11. . #etrieved 2010-08-22.
O "It's probable that you've misunderstood 'Cloud Computing' until now" (http:/ / portal.
acm. org/ citation. cIm?id1496091. 1496100& coll& dlACM& CFID21518680&
CFTOKEN18800807). TechPluto. . #etrieved 2010-09-14.
O "Service-Oriented Computing and Cloud Computing: Challenges and Opportunities"
(http:/ / www. computer. org/ portal/ web/ csdl/ doi/ 10. 1109/ MIC. 2010. 147). IEEE
Internet Computing. . #etrieved 2010-12-04.