15CS565 CLOUD COMPUTING

Module - 5
Cloud Applications. ( Chapter - 10 )
Cloud computing has gained huge popularity in industry due to its ability to host applications for which the
services can be delivered to consumers rapidly at minimal cost.
This chapter discusses some application case studies, detailing their architecture and how they leveraged
various cloud technologies.
Applications from a range of domains, from scientific to engineering, gaming, and social networking, are
considered.

10.1 Scientific applications

10.1.1 Healthcare: ECG analysis in the cloud
10.1.2 Biology: protein structure prediction
10.1.3 Biology: gene expression data analysis for cancer diagnosis
10.1.4 Geoscience: satellite image processing
10.2 Business and consumer applications
10.2.1 CRM and ERP
1 Salesforce.com
2 Microsoft dynamics CRM
3 NetSuite
10.2.2 Productivity
1 Dropbox and iCloud
2 Google docs
3 Cloud desktops: EyeOS and XIOS/3
10.2.3 Social networking
1 Facebook
10.2.4 Media applications
1 Animoto
2 Maya rendering with Aneka
3 Video encoding on the cloud: Encoding.com
10.2.5 Multiplayer online gaming

10.1 Scientific Applications

Scientific applications are a sector that is increasingly using cloud computing systems and technologies.
Cloud computing systems meet the needs of different types of applications in the scientific domain:
high-performance computing (HPC) applications, high-throughput computing (HTC) applications, and
data-intensive applications.
The opportunity to use cloud resources is even more appealing because minimal changes need to be made to
existing applications in order to leverage cloud resources.

10.1.1 Healthcare: ECG analysis in the cloud

Healthcare is a domain in which computer technology has found several and diverse applications: from
supporting the business functions to assisting scientists in developing solutions to cure diseases.
An illustration of the infrastructure and model for supporting remote ECG monitoring is shown in Figure 10.1.
Wearable computing devices equipped with ECG sensors constantly monitor the patients heartbeat. Such
information is transmitted to the patients mobile device, which will eventually forward it to the cloud-hosted
Web service for analysis.

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 1

15CS565 CLOUD COMPUTING

The Web service forms the front-end of a platform that is hosted in cloud and leverages three layers of cloud
computing stack: SaaS, PaaS, and IaaS.
The Web service constitute SaaS application that will store ECG data in the Amazon S3 service and issue a
processing request to the scalable cloud platform.
The runtime platform is composed of a dynamically sizable number of instances running the workflow engine
and Aneka.
The number of workflow engine instances is controlled according to the number of requests in the queue of
each instance, while Aneka controls the number of EC2 instances used to execute the single tasks defined by
the workflow engine for a single ECG processing job.
Advantages
1. The elasticity of cloud infrastructure that can grow and shrink according to the requests served. As a result,
doctors and hospitals do not have to invest in large computing infrastructures designed after capacity planning,
thus making more effective use of budgets.
2. Ubiquity. Cloud computing technologies are easily accessible and promise to deliver systems with minimum
or no downtime. Computing systems hosted in cloud are accessible from any Internet device through simple
interfaces (such as SOAP and REST-based Web services). This makes systems easily integrated with other
systems maintained on hospitals premises.
3. Cost savings. Cloud services are priced on a pay-per-use basis and with volume prices for large numbers of
service requests.

10.1.2 Biology: protein structure prediction

Applications in biology require high computing capabilities and operate on large data-sets that cause extensive
I/O operations.
Therefore biology applications have made use of supercomputing and cluster computing infrastructures.
Similar capabilities can be leveraged using cloud computing technologies in a more dynamic fashion, thus
opening new opportunities for bioinformatics applications.

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 2

15CS565 CLOUD COMPUTING

Protein structure prediction is a computationally intensive task that is fundamental to different types of research
in the life sciences.
The geometric structure of a protein cannot be directly inferred from the sequence of genes that compose its
structure, but it is the result of complex computations aimed at identifying the structure that minimizes the
required energy.
This task requires the investigation of a space with a massive number of states, consequently creating a large
number of computations for each of these states.
One project that investigates the use of cloud technologies for protein structure prediction is Jeeva - an
integrated Web portal that enables scientists to offload the prediction task to a computing cloud based on Aneka
(Figure 10.2).

The prediction task uses machine learning techniques (support vector machines) for determining the secondary
structure of proteins.
These techniques translate problem into one of pattern recognition, where a sequence has to be classified into
one of three possible classes (E, H, and C).
A popular implementation based on support vector machines divides the pattern recognition problem into three
phases: initialization, classification, and a final phase.
These three phases have to be executed in sequence, we can perform parallel execution in the classification
phase, where multiple classifiers are executed concurrently.
This reduces computational time of the prediction.
The prediction algorithm is then translated into a task graph that is submitted to Aneka.
Once the task is completed, the middleware makes the results available for visualization through the portal.
The advantage of using cloud technologies is the capability to leverage a scalable computing infrastructure
that can be grown and shrunk on demand.

10.1.3 Biology: gene expression data analysis for cancer diagnosis

Gene expression profiling is the measurement of the expression levels of thousands of genes at once. It is used
to understand the biological processes that are triggered by medical treatment at a cellular level.
important application of gene expression profiling is cancer diagnosis and treatment.
Cancer is a disease characterized by uncontrolled cell growth and proliferation. This behavior occurs because
genes regulating the cell growth mutate.
Gene expression profiling is utilized to provide a more accurate classification of tumors.

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 3

15CS565 CLOUD COMPUTING

The dimensionality of typical gene expression datasets ranges from several thousands to over tens of thousands
of genes.
This problem is approached with learning classifiers, which generate a population of condition-action rules that
guide the classification process. The eXtended Classifier System (XCS) has been utilized for classifying large
datasets in bioinformatics and computer science domains.
A variation of this algorithm, CoXCS [162], has proven to be effective in these conditions. CoXCS divides the
entire search space into subdomains and employs the standard XCS algorithm in each of these subdomains.
Such a process is computationally intensive but can be easily parallelized because the classifications problems
on the subdomains can be solved concurrently.
Cloud-CoXCS (Figure 10.3) is a cloud-based implementation of CoXCS that leverages Aneka to solve the
classification problems in parallel and compose their outcomes. The algorithm is controlled by strategies, which
define the way the outcomes are composed together and whether the process needs to be iterated.

10.1.4 Geoscience: satellite image processing

Geoscience applications collect, produce, and analyze massive amounts of geospatial and nonspatial data. As
the technology progresses and our planet becomes more instrumented, volume of data that needs to be
processed increases significantly.
Geographic information system (GIS) applications capture, store, manipulate, analyze, manage, and present all
types of geographically referenced data.
Cloud computing is an attractive option for executing these demanding tasks and extracting meaningful
information to support decision makers.
Satellite remote sensing generates hundreds of gigabytes of raw images that need to be further processed to
become the basis of several different GIS products. This process requires both I/O and compute-intensive tasks.
Large images need to be moved from a ground stations local storage to compute facilities, where several
transformations and corrections are applied.
The system shown in Figure 10.4 integrates several technologies across the entire computing stack.
A SaaS application provides a collection of services for such tasks as geocode generation and data
visualization.
At the PaaS level, Aneka controls the importing of data into the virtualized infrastructure and the execution of
image-processing tasks that produce the desired outcome from raw satellite images.
The platform leverages a Xen private cloud and the Aneka technology to dynamically provision the required
resources.

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 4

15CS565 CLOUD COMPUTING

10.2 Business and consumer applications

The business and consumer sector is the one that benefits the most from cloud computing technologies.
On one hand, the opportunity to transform capital costs into operational costs makes clouds an attractive option
for all enterprises that are IT-centric.
On the other hand, the sense of ubiquity that the cloud offers for accessing data and services makes it
interesting for end users.
The combination of all these elements has made cloud computing the preferred technology for a wide range of
applications.

10.2.1 CRM and ERP

Customer relationship management (CRM) and enterprise resource planning (ERP) applications are market
segments that are flourishing in the cloud
Cloud CRM applications constitute a great opportunity for small enterprises and start-ups to have fully
functional CRM software without large up-front costs and by paying subscriptions.
Your business and customer data from everywhere and from any device, has fostered the spread of cloud CRM
applications.
ERP solutions on the cloud are less mature and have to compete with well-established in-house solutions. ERP
systems integrate several aspects of an enterprise: finance and accounting, human resources, manufacturing,
supply chain management, project management, and CRM.

1 Salesforce.com
Salesforce.com is most popular and developed CRM solution available today.
As of today more than 100,000 customers have chosen Safesforce.com to implement their CRM solutions.
The application provides customizable CRM solutions that can be integrated with additional features developed
by third parties.
Salesforce.com is based on the Force.com cloud development platform.
This represents scalable and high-performance middleware executing all operations of all Salesforce.com
applications.
The architecture of the Force.com platform is shown in Figure 10.5.

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 5

15CS565 CLOUD COMPUTING

At the core of the platform resides its metadata architecture, which provides the system with flexibility and
scalability.
Application core logic and business rules are saved as metadata into the Force.com store.
Both application structure and application data are stored in the store. A runtime engine executes application
logic by retrieving its metadata and then performing the operations on the data.
A full-text search engine supports the runtime engine. This allows application users to have an effective user
experience The search engine maintains its indexing data in a separate store.

2 Microsoft dynamics CRM

The system is completely hosted in Microsofts datacenters across the world and offers to customers a 99.9%
SLA.
Each CRM instance is deployed on a separate database, and application provides users with facilities for
marketing, sales, and advanced customer relationship management.
Dynamics CRM Online features can be accessed either through a Web browser interface or by means of SOAP
and RESTful Web services.
This allows Dynamics CRM to be easily integrated with both other Microsoft products and line-of-business
applications.
Dynamics CRM can be extended by developing plug-ins that allow implementing specific behaviors triggered
on the occurrence of given events.

3 NetSuite
NetSuite provides a collection of applications that help customers manage every aspect of the business
enterprise.
Its offering is divided into three major products: NetSuite Global ERP, NetSuite Global CRM1 , and NetSuite
Global Ecommerce.
Moreover, an all-in-one solution: NetSuite One World, integrates all three products together.

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 6

15CS565 CLOUD COMPUTING

The services NetSuite delivers are powered by two large datacenters on the East and West coasts of the United
States, connected by redundant links.
This allows NetSuite to guarantee 99.5% uptime to its customers.
The NetSuite Business Operating System (NS-BOS) is a complete stack of technologies for building SaaS
business applications that leverage the capabilities of NetSuite products.
On top of the SaaS infrastructure, the NetSuite Business Suite components offer accounting, ERP, CRM, and
ecommerce capabilities.

10.2.2 Productivity
Productivity applications replicate in cloud. The most common tasks that we are used to performing on our
desktop: from document storage to office automation and complete desktop environments hosted in the cloud.
1 Dropbox and iCloud
Online storage solutions have turned into SaaS applications and become more usable as well as more advanced
and accessible.
The most popular solution for online document storage is Dropbox, that allows users to synchronize any file
across any platform and any device in a seamless manner (Figure 10.6). Dropbox provides users with a free
storage that is accessible through the abstraction of a folder. Users can either access their Dropbox folder
through a browser or by downloading and installing a Dropbox client, which provides access to the online
storage by means of a special folder. All the modifications into this folder are silently synched so that changes
are notified to all the local instances of the Dropbox folder across all the devices.
Another interesting application in this area is iCloud, a cloud-based document-sharing application provided by
Apple to synchronize iOS-based devices in a completely transparent manner.
Documents, photos, and videos are automatically synched as changes are made, without any explicit operation.
This allows the system to efficiently automate common operations without any human intervention.
This capability is limited to iOS devices, and currently there are no plans to provide iCloud with a Web-based
interface that would make user content accessible from even unsupported platforms.

2 Google docs
Google Docs is a SaaS application that delivers the basic office automation capabilities with support for
collaborative editing over the Web.
Google Docs allows users to create and edit text documents, spreadsheets, presentations, forms, and drawings.
It aims to replace desktop products such as Microsoft Office and OpenOffice and provide similar interface and
functionality as a cloud service.
By being stored in the Google infrastructure, these documents are always available from anywhere and from
any device that is connected to the Internet.

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 7

15CS565 CLOUD COMPUTING

Google Docs is a good example of what cloud computing can deliver to end users: ubiquitous access to
resources, elasticity, absence of installation and maintenance costs, and delivery of core functionalities as a
service.

3 Cloud desktops: EyeOS and XIOS/3

EyeOS is one of the most popular Web desktop solutions based on cloud technologies. It replicates the
functionalities of a classic desktop environment and comes with pre-installed applications for the most common
file and document management tasks (Figure 10.7). Single users can access the EyeOS desktop environment
from anywhere and through any Internet-connected device, whereas organizations can create a private EyeOS
Cloud on their premises to virtualize the desktop environment of their employees and centralize their
management.

The EyeOS architecture is quite simple: On the server side, the EyeOS application maintains the information
about user profiles and their data, and the client side constitutes the access point for users and administrators to
interact with the system. EyeOS stores the data about users and applications on the server file system. Once the
user has logged in by providing credentials, the desktop environment is rendered in the clients browser by
downloading all the JavaScript libraries required to build the user interface and implement the core
functionalities of EyeOS.
EyeOS also provides APIs for developing new applications and integrating new capabilities into the system.
EyeOS applications are server-side components that are defined by at least two files (stored in the
eyeos/apps/appname directory): appname.php and appname.js. The first file defines and implements all the
operations that the application exposes; the JavaScript file contains the code that needs to be loaded in the
browser in order to provide user interaction with the application.
Xcerion XML Internet OS/3 (XIOS/3) is another example of a Web desktop environment. The service is
delivered as part of the CloudMe application, which is a solution for cloud document storage. The key
differentiator of XIOS/3 is its strong leverage of XML, used to implement many of the tasks of the OS:
rendering user interfaces, defining application business logics, structuring file system organization, and even
application development.
XIOS/3 is released as open-source software and implements a marketplace where third parties can easily deploy
applications that can be installed on top of the virtual desktop environment. It is possible to develop any type of

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 8

15CS565 CLOUD COMPUTING

application and feed it with data accessible through XML Web services: developers have to define the user
interface, bind UI components to service calls and operations, and provide the logic on how to process the data.

10.2.3 Social networking

Social networking applications have grown considerably in the last few years to become the most active sites
on the Web.
To sustain their traffic and serve millions of users seamlessly, services such as Twitter and Facebook have
leveraged cloud computing technologies.

1 Facebook
Facebook is probably the most evident and interesting environment in social networking.
With more than 800 million users, it has become one of the largest Websites in the world.
To sustain this incredible growth, it has been fundamental that Facebook be capable of continuously adding
capacity and developing new scalable technologies and software systems while maintaining high performance
to ensure a smooth user experience.
Currently, the social network is backed by two data centers that have been built and optimized to reduce costs
and impact on the environment.
On top of this highly efficient infrastructure, built and designed out of inexpensive hardware, a completely
customized stack of opportunely modified and refined open-source technologies constitutes the back-end of the
largest social network.
The reference stack serving Facebook is based on LAMP (Linux, Apache, MySQL, and PHP). This collection
of technologies is accompanied by a collection of other services developed in-house.
These services are developed in a variety of languages and implement specific functionalities such as search,
news feeds, notifications, and others.
While serving page requests, the social graph of the user is composed.
The social graph identifies a collection of interlinked information that is of relevance for a given user.
Most of the user data are served by querying a distributed cluster of MySQL instances, which mostly contain
key-value pairs.

10.2.4 Media applications

Media applications has taken a considerable advantage from leveraging cloud computing technologies.
The computationally intensive tasks can be easily offloaded to cloud computing infrastructures.

1 Animoto
Animoto is the most popular example of media applications on the cloud. The Website provides users with a
very straightforward interface for quickly creating videos out of images, music, and video fragments submitted
by users. Users select a specific theme for a video, upload the photos and videos and order them in the sequence
they want to appear, select the song for the music, and render the video. The process is executed in the
background and the user is notified via email once the video is rendered.
A proprietary artificial intelligence (AI) engine, which selects the animation and transition effects according to
pictures and music, drives the rendering operation. Users only have to define the storyboard by organizing
pictures and videos into the desired sequence.
The infrastructure of Animoto is complex and is composed of different systems that all need to scale ( Figure
10.8). The core function is implemented on top of the Amazon Web Services infrastructure. It uses Amazon
EC2 for the Web front-end and worker nodes; Amazon S3 for the storage of pictures, music, and videos; and
Amazon SQS for connecting all the components.

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 9

15CS565 CLOUD COMPUTING

The systems auto-scaling capabilities are managed by Rightscale, which monitors the load and controls the
creation of new worker instances.

2 Maya rendering with Aneka

A private cloud solution for rendering train designs has been implemented by the engineering department of
GoFront group, a division of China Southern Railway (Figure 10.9). The department is responsible for
designing models of high-speed electric locomotives, metro cars, urban transportation vehicles, and motor
trains. The design process for prototypes requires high-quality, three-dimensional (3D) images. The analysis of
these images can help engineers identify problems and correct their design.
Three-dimensional rendering tasks take considerable amounts of time, especially in the case of huge numbers
of frames, but it is critical for the department to reduce the time spent in these iterations. This goal has been
achieved by leveraging cloud computing technologies, which turned the network of desktops in the department
into a desktop cloud managed by Aneka.

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 10

15CS565 CLOUD COMPUTING

3 Video encoding on the cloud: Encoding.com

Video encoding and transcoding are operations that can greatly benefit from using cloud technologies: They are
computationally intensive and potentially require considerable amounts of storage.
Encoding.com is a software solution that offers video-transcoding services on demand and leverages cloud
technology to provide both the horsepower required for video conversion and the storage for staging videos.
The service integrates with both Amazon Web Services technologies (EC2, S3, and CloudFront) and Rackspace
(Cloud Servers, Cloud Files, and Limelight CDN access).
To use the service, users have to specify the location of the video to transcode, the destination format, and the
target location of the video. Encoding.com also offers other video-editing operations such as the insertion of
thumbnails, watermarks, or logos. Moreover, it extends its capabilities to audio and image conversion.

10.2.5 Multiplayer online gaming

Online multiplayer gaming attracts millions of gamers around the world who share a common experience by
playing together in a virtual environment that extends beyond the boundaries of a normal LAN. Online games
support hundreds of players in the same session, made possible by the specific architecture used to forward
interactions, which is based on game log processing.
Players update the game server hosting the game session, and the server integrates all the updates into a log that
is made available to all the players through a TCP port. The client software used for the game connects to the
log port and, by reading the log, updates the local user interface with the actions of other players.

Game log processing is also utilized to build statistics on players and rank them. These features constitute the
additional value of online gaming portals that attract more and more gamers. The processing of game logs is a
potentially compute-intensive operation that strongly depends on the number of players online and the number
of games monitored.
The use of cloud computing technologies can provide the required elasticity for seamlessly processing these
workloads and scale as required when the number of users increases. A prototype implementation of
cloud-based game log processing has been implemented by Titan Inc. (Figure 10.10)

Prof. Altaf Husain A. Makandar, Dept. of CSE, AGMRCET, Hubli 11