Unit-1 Computing Paradigms 1.

Marketed by:

Unit Computing Paradigms

1 SIA GROUP

Part-A
Short Questions with Solutions
Q1. Define parallel computing.
Answer : Model Paper-III, Q1(a)

The computing wherein a single program is executed by multiple processors simultaneously is referred to as parallel
computing. The major advantage of parallel computing is that, the execution of a program is done at very high speed. However,
the difficulty of such computing lies in dividing a single program among multiple processors without interfering with one another.
This sort of computing is generally performed on a computer that contain more than one process. However, it can also be performed
by connecting computers into a network. This computing is specially used in areas such as weather forecasting and semiconductor
design. Beside this, parallel computing is also used to solve such problem which cannot be solved by single computer.
Q2. Define cluster computing.
Answer : Model Paper-II, Q1(a)

Cluster is a term referring to a collection of computers connected over a local network and work together as a single unified
resource. Cluster computing is a mechanism of combining these multiple computers together to work as a single high-performance
computer. A cluster computer can be used as an alternative to super computer because it provides improved performance in terms
of speed on a wide range of high-performance computing applications.
Q3. What are the advantages of grid computing?
Answer : Model Paper-I, Q1(b)

The advantages of grid computing are as follows,

v Availability of high computational power.
v Provide access to certain additional resources other than processor and storage.
v Large storage capacity.
v As the resources are distributed geographically, failure of one resource will not effect the functionality of others.
v Resources can be distributed based on CPU utilization of nodes.
Q4. Write a short note on Biocomputing.
Answer : Model Paper-III, Q1(b)

The biocomputing systems operate on the idea of biologically derived or simulated molecules. This is necessary for
carryingout computational processes for solving problems. Moreover, the biologically derived models assist in structuring the
computer programs to include in the application.
The biocomputing not just caters theoretical background but also caters practical tools. This is helpful for scientists for
carryingout research work in protein patterns and DNA. In simple terms, the DNA and proteins can be regarded as building blocks
and its functionality is highly dependant on the order of these blocks.

SIA Group
1.2 Cloud Computing [JNTU-hyderabad]
Q5. Write about Quantum Computing.

Answer : Model Paper-I, Q1(a)

The quantum computing came into existence inorder to solve the problem of manufacturers of computing systems. The
problem is embedding of more and more transistors into tiny spaces of integrated circuits. This addition demands equivalent
processing power i.e., the processing power should be doubled in every 18 months. So, the new quantum computing based solution
comes into picture where the entire focus is made on quantum information.

Q6. Write a short note on Network Computing.

Answer : Model Paper-II, Q1(b)

The network computing can be viewed as a method of developing systems. This assist in using benefits of latest technology
and increasing the positive impact on the business solutions. As a consequence, the ability of these systems gets enhanced and
they serve better to the customers because they operate on strong network of computing resources.
In most of the network computing solution, the client entity corresponding to networked architecture or application will
support customer or client or end-user. But these days, they offer necessary set of functionalities which sustains proper client
functions. This is accomplished at minimum cost by keeping it very simple. On the other hand, they are different from conventional
PCs and does not require any individual configuration. Also, their maintenance is done based upon the requirements.
The server will be the next entity of the component with in the network architecture. They transfer the application services
the client end. Therefore, all the computing paradigms follow this architecture and it is possible that the newly developed paradigms
will be designed based on network architecture.

L o o k f o r t h e SIA Group logo o n t h e title cover before you buy

Unit-1 Computing Paradigms 1.3

Part-B
ESSAY Questions with Solutions

1.1 High-Performance Computing, Parallel Computing, Distributed

Computing
Q7. Discuss in detail about High-Performance computing.

Answer : Model Paper-I, Q2(a)

High-Performance Computing

HPC stands for High-Performance Computing which improves the performance of computing by increasing its speed. The
main motivation behind such kind of computing is improvements in various fields such as engineering, scientific etc., which made
the speed measured over Pflops from Gflops. Modern computing usually involve desktop (or) PCs are connected over LANs and
WANs where computing must provide efficient ways to make the servers transparent along with faster access.

The high-performance computing systems are typically a set of processors which could be processor machines or CPUs.
These systems are connected together through devices like memory unit, storage unit and input or output devices. Consecutively,
the entire setup of connected components is run using the software which is employed.

The HPC can include both same type of processor machines or different types of processor machines. Since, they attribute
high-performance. So, they are sometimes called as super computers and are difficult to replace. They also include computing
paradigms.

Some of the examples of HPC could be, small cluster of desktop computers, Personal Computers (PCs) and fastest super
computers. HPC systems are predominantly used in applications which include scientific problem solving. In essence, these
problems demands appropriate simulation study which can easily be accomplished through HPC. Typical scientific problems
include protein folding in molecular biology, research and study on developing models and applications depending on nuclear
fusion.
Q8. Write about parallel computing.

Answer : Model Paper-II, Q2(a)

The computing wherein a single program is executed by multiple processors simultaneously is referred to as parallel
computing. The major advantage of parallel computing is that, the execution of a program is done at very high speed. However,
the difficulty of such computing lies in dividing a single program among multiple processors without interfering with one
another. This sort of computing is generally performed on a computer that contain more than one process. However, it can also
be performed by connecting computers into a network. This computing is specially used in areas such as weather forecasting
and semiconductor design. Beside this, parallel computing is also used to solve such problem which cannot be solved by single
computer.

It is one of the aspect of HPC. This computing collectively solves a computational problem through systems which are
of same type (i.e., same type of CPUs). In a way similar to HPC, parallel computing too includes super computers consisting of
hundreds and thousands of processors connected to other resources.

The main difference between serial (or sequential computers) and parallel computing are discussed below,

The former one runs on single computer which utilizes only single CPU. It disintegrates the problem into different series
of instructions. Here, the execution of instructions is done in sequential manner. While the latter one use many processors at
the same time using multiple CPUs. Here, the problem is disintegrated into different portions and can be solved at the same
time. And also, every portion is further divided into sequence of instructions, which undergo concurrent execution on multiple
processors.

SIA Group
1.4 Cloud Computing [JNTU-hyderabad]

Q9. Discuss in brief about distributed computing.

Answer : Model Paper-III, Q2(a)

Distributed Computing

In the distributed computing model, the processing is done in multiple computers that are connected in the same network.
Each of these computers have their own processors in addition to other resources. The overall structure of a distributed computing
model is as shown in figure.

Figure: Distributed Computing Model

All the workstations, i.e., nodes have a complete access on the resources of local computer to which it is connected. When
there is an interaction between local computers and remote computers, the users will be able to access the resources of remote
computers as well.

A best example of distributed computing is the ‘World Wide Web’. To visit a website, a user makes use of a browser for
example Internet Explorer, Mozilla Firefox or Netscape Navigator etc., They will run on local system of the user. They will also
interact with another program running on a remote system to search for a required file. This file may reside on any other remote
system.

The distributed computing network consists of either same system or different systems. But irrespective of this, it runs on
single system. The CPUs in this network could physically be close together or can be connected through local network. It is also
possible that, the CPUs can geographically be kept at some distance and connected through wide area network. Due to the inclusion
of heterogeneity (i.e., different systems), the distributed computing network sustains any number of possible configurations with
in the processor machines. These processor machines could be mainframes, PCs, workstations and minicomputers. Thus, making
the entire set-up as single computer.

The distributed systems substantiates many characteristic features in comparison to centralized systems. Some of
them are,

1. Scalability

The distributed systems are highly scalable and can be expanded by including more machines and vice-versa. Also, the
addition of the machines does not disturbs the existing set-up.

2. Redundancy or Replication

The distributed systems support the feature of redundancy where in many computers provide same services. This feature
allows the system to function even if any of the computer fails. This is because of the availability of other similar computing
supports.

L o o k f o r t h e SIA Group logo o n t h e title cover before you buy

Unit-1 Computing Paradigms 1.5

1.2 Cluster Computing, Grid Computing, Cloud Computing

Q10. Write about the following,

(i) Cluster Computing
(ii) Grid Computing.
Answer : Model Paper-I, Q3

(i) Cluster Computing

Cluster is a term referring to a collection of computers connected over a local network and work together as a single unified
resource. Cluster computing is a mechanism of combining these multiple computers together to work as a single high-performance
computer. A cluster computer can be used as an alternative to super computer because it provides improved performance in terms
of speed on a wide range of high-performance computing applications.
These clusters consist of identical set of processor machines well connected through dedicated network infrastructure. These
processor machines permits the programs to be run across all nodes concurrently. This is possible, because processor machines
shares resources such as common home directory and holds a software called Message Passing Interface (MPI) implementation
to substantiate it.
Ideally, cluster computing is also one of the aspect of HPC category. Here, all individual nodes operates together through
meaningful communication with each other so as to solve complex problems.
Advantages
v It incurs low cost when compared with supercomputers.
v As multiple computers (or) components work together as a single computer, it can be easily managed.
v It uses replicas of various parts of the system which makes the resources highly available.
v It can be extended easily if additional computational power is required.
Disadvantages
v It does not provide the feature of resource sharing among participants of different clusters.
v Resources belonging to different software background might generate issues while combining.
v As there exist replicas of various components, finding faulty component is a tedious task.
(ii) Grid Computing
A computational grid is typically referred to a group of computing resources that belong to individual, heterogeneous
(having different operating systems) and geographically different computers that carry out a common goal. These computers are
loosely coupled through a network and usually shares their idle CPU cycles with each other. The cycles from all participants of
a grid together forms a ‘Virtual Super Computers’.
Here, a large task is divided into multiple smaller tasks and distributed among all the computers of a grid. This distribution
is typically carried out with the help of middleware. The maximum size of a grid might be equal to the total number of computer
participating in a grid.
The term grid was generally used to refer electric power grid but in computing terms, it was first used as power grid metaphor
in the beginning of 1990s. Ian Foster, Steve Tuecke and Calrl Kesselmen are referred to as fathers of grid because they included
features like distributed computing, web services and object oriented programming in the computational grid. They introduced
Globus toolkit that was considered as a major solution to many of the computing problems.
Many computing resources of organizations are underutilized and kept idle mostly. They are necessary for certain operations.
So, the concept of grid computing was brought into focus to utilize such underutilized computing power and supply it to the needy
organizations. This can eventually leads to increase in the ReturnOnInvestment (ROI) on computing investments.
Furthermore, the software required is middleware which allows the remote access of resources. Each activity of grid
resources managed through middleware is called grid services. Some of its services include access control, security, data access
which includes digital libraries and databases. It also provides access to large-scale interactive and long-term storage facilities.

SIA Group
1.6 Cloud Computing [JNTU-hyderabad]
The grid computing is highly popular because of its ability to utilize unused computing power. It is also a cost-
effective solution to many computing problems. It holds many options to alleviate problems in regard to any HPC based
application.

It also allow heterogeneous resources to operate cooperatively and collaboratively for solving any complicated scientific
problem.

Advantages

v Availability of high computational power.

v Provide access to certain additional resources other than processor and storage.

v Large storage capacity.

v As the resources are distributed geographically, failure of one resource will not effect the functionality of others.

v Resources can be distributed based on CPU utilization of nodes.

Disadvantages

v It does not provide stable standards.

v Combining resources from geographically distributed sites require high speed internet connectivity.

v Efficient tools such as opsware are needed to perform operations like synchronization.

Q11. Discuss in detail about cloud computing and its history.

Answer : Model Paper-III, Q2(b)

Cloud Computing

The term cloud is usually used to represent the Internet but it is not just restricted to Internet. It is a virtual storage where
the data is stored in third-party data centers. Storing, managing and accessing data present in the cloud is typically referred to
as cloud computing. It is a model for distributing information technology inorder to gain access to the resources from Internet
without depending on direct connection with the server. It uses various web-based tools and applications to easily retrieve
resources.

Accessing resources over the Internet makes these resources available anytime and anywhere thereby allowing the users
to work remotely. In general, cloud computing is nothing but the use of computing resources such as hardware and software that
are distributed as services across the network. It centralizes the data storage, processing and bandwidth which in turn provides
efficient computing to the user. The services are made available by a cloud vendor based on pay-per-use.

In order to serve large computing resources for solving single problem, the concept of computing escalated from grid
computing to cloud computing. This computing make use of potential ideas of computing power in the form of utility. The main
difference between grid and cloud is that, the former one substantiate the use of multiple computers concurrently for solving a
specific application. On the other hand, the cloud computing substantiates the use of multiple resources which includes computing
resources inorder to serve unified service to the end user.

Typically, the cloud computing holds IT and business resources including servers storage, network, applications and
processes. It provides the user needs and workload dynamically. Apart from supporting grid, the cloud also supports non-grid
environments including three tier web architecture.

History of Cloud Computing

In 1977, the figure of cloud was used in ARPANET to represent network of equipments associated with it. A few years
later, it was used to represent platforms associated with distributed systems. In 2006, it became popular as Amazon released EC2
(Elastic Computer Cloud) as a commercial web service.

L o o k f o r t h e SIA Group logo o n t h e title cover before you buy

Unit-1 Computing Paradigms 1.7

Practically, the cloud computing has evolved through various phases among which grid and utility computing holds much
of its significance. The history of cloud computing is discussed below,
1970s
In 1970s, most of the systems were based on time-sharing terminologies among which, the popular one was Remote Job
Entry (RJE). In 1972, IBM introduced its first virtual machine operating system that makes use of various virtual machines to
provide a time-sharing operational services.
1990s
With introduction of the concept of virtual machines, most of the organizations that used to have point-to-point data
circuits started using VPNs (Virtual Private Networks) which incurs low cost. Moreover, they were able to manage the network
traffic more efficiently. At this point, telecommunication organizations started using cloud symbol to represent interface between
customer premises equipment (such as wiring) and service provider equipment. However, cloud computing is much more beyond
this boundary and includes all types of servers.
In 1999, the introduction of salesforce.com lead to the concept of providing services and applications over various websites.
2000
In 2002, Amazon started providing web services such as storage, computations etc., with the introduction of Amazon
Mechanical Turk. A few years later, in 2006, Amazon introduced a much more effective mechanism of providing commercial
webservices on rental basis with the use of Amazon EC2 (Elastic Computer Cloud).
In 2008, NASA introduced the first-open source software capable of supporting both hybrid and private cloud but lacks in
providing quality of service. In the same year, microsoft introduced “Azure” which was included in the top 500 supercomputers.
In 2011, IBM introduced IBM smart cloud which is a framework used to provide assistance in smarter planet. Later on in
2012, Oracle developed oracle cloud which is capable of providing interesting cloud services such as SaaS (Software as a Service),
PaaS (Platform as a Service) and IaaS (Infrastructure as a Service).
Q12. State the basic principles of cloud computing.
Answer : Model Paper-II, Q3(a)

The following are the basic principles of cloud computing,

1. Federation
A cloud computing environment must be capable of providing federated service providers which means that, these providers
must be capable of collaborating and resource sharing at any point irrespective of their type. This is usually needed when
an organization extends its computing paradigm from private to public cloud. Moreover, this federation must be kept
transparent so that the virtual application can be used on all the sites. This makes the application to be handled remotely
and allows it to migrate from one site to another. Apart from this, federation must be carried out in a secure and independent
way.
2. Independence
The user of cloud computing services must be independent of the providers specific tool and the type of service. According
to this principle, a user must be allocated with required virtual resource irrespective of the type of provider. Moreover, it
is the responsibility of service providers to handle cloud infrastructure while hiding the confidential information.
3. Isolation
According to this principle, a service provider must ensure the user with respect to the isolation of their data from others.
Even the data in the same cloud must be separated from different users and therefore should not be accessed.
4. Elasticity
The users of cloud computing must be providing with ease of accessing and releasing the resources as required. This
typically referred to as elasticity. The rules associated with elasticity must be included within the contract made between
consumers and service providers.

SIA Group
1.8 Cloud Computing [JNTU-hyderabad]

5. Business Orientation
To develop a more efficient computing environment, an efficient platform must be developed before the applications are
included into the cloud. This typically ensure the quality of service and assist SLAs (Service-Level Agreements).
6. Trust
To build a successful cloud computing environment, one of the major factor is trust between consumers and service
providers. Therefore, effective mechanism must be included to develop trust worthy computing environment.
Q13. List the features of cloud computing.
Answer :
The key features of cloud computing are as follows,
1. Low Capital
It offers services to specific consumers and small business because these services help them to minimize their capital
expenditure.
2. Convenience
It allows the customer to access the cloud computing services regardless of their locations or type of devices.
3. Resource Sharing
It employs multitendancy concept that allows sharing of resources and cost among large number of users. Due to this, the
efficiency of under utilize systems increases along with the peak load capacity.
4. Reliability
It provides high reliability because service providers make use of multiple sites.
5. Scalability
It offers high scalability depending upon the fluctuating demands of the users.
6. Security
(i) It provides high level of security to the customer resources and infrastructure which is maintained by service providers.
(ii) It employs dedicated security staff and protection policies inorder to keep the customer data safe and secure.
7. Authorization
It employs data protection policies and access to data is provided only to authenticated users. However, the security
mechanisms used for audit log are extreme which does not general many customers to access it.
Q14. List the challenges of cloud computing.
Answer : Model Paper-II, Q3(b)

The challenges faced by the companies that are implementing cloud computing are as follows,
1. Secured data storage
2. High speed internet access
3. Standardization.
4. Availability, fault tolerance and disaster recovery
5. Resources management and efficiency.
1. Secured Data Storage
Organizations stores their large amount of confidential data outside the corporate firewalls at centralised locations. The data
could be corresponding to user privacy, identity and application specific preferences. But the data protection mechanisms
across these locations is likely to be compromised by Hackers and various attacks. Due to which the overall system
performance decreases. This may in turn create various doubts/confusions regarding the legal frame work that the cloud
computing has adopted.

L o o k f o r t h e SIA Group logo o n t h e title cover before you buy

Unit-1 Computing Paradigms 1.9
2. High Speed Internet Access
The organizations can exercise cloud computing provided if it has constant access to high speed internet. Although, the
organizations perform cost cutting over hardware, they have to spend significant amount for data intensive applications
as they require sufficient bandwidth. Another important concern is the fact that, the internet broad-band service in United
States is very poor in comparison to other European and Asian countries. Therefore, cloud computing and its services
cannot be pursuit unless the organization has access to high-speed internet. So, these challenges must be overcome for the
betterment of the organization.
3. Standardization
An organization in order to exercise cloud computing, must adopt high-level of technical standards. These technical
standards is necessary for implementing various computer systems and applications. However, these standards has still not
been defined nor introduced in public and not even approved by external body. Hence, these hurdles inhibits the growth
of an organization.
Apart from these challenges, the cloud computing also faces the challenge of reliability. The organization even though uses
the computing services, often get confused regarding the reliability of the services, as there is a threat of data being stolen,
power failures, viruses, hackers. The cloud computing services are also not robust in nature due to which the performance
may degrade.
4. Availability, Fault Tolerance and Disaster Recovery
The users of cloud services expect that their services and resources must be available any time and anywhere even if errors
(faults) exist in the system. Moreover, the system must include the feature of recovering from disasters that might occur
for any reason.
5. Resources Management and Efficiency
As the resources such as CPU cores, disk space etc are distributed and shared among multiple machines/networks, managing
these resources is considered as a major challenge. Moreover, operations like load balancing, backup and recovery must
be performed without degrading the performance of the system.

1.3 Bio Computing, Mobile Computing, Quantum Computing, Optical

Computing, Nano Computing

Q15. Briefly discuss about biocomputing.

Answer : Model Paper-II, Q2(b)

The biocomputing systems operate on the idea of biologically derived or simulated molecules. This is necessary for
carryingout computational processes for solving problems. Moreover, the biologically derived models assist in structuring the
computer programs to include in the application.
The biocomputing not just caters theoretical background but also caters practical tools. This is helpful for scientists for
carryingout research work in protein patterns and DNA. In simple terms, the DNA and proteins can be regarded as building blocks
and its functionality is highly dependant on the order of these blocks.
Therefore, the biocomputing scientist operates on inventing the order needed for various application. This entails better
comprehension of life and molecular causes of diseases.
Q16. Write a short note on mobile computing.
Answer : Model Paper-I, Q2(b)

Mobile Computing
The term ‘Mobile’ is the ability to be on the move. Mobile computing systems are the computing systems that may be
easily moved while they are being moved. Examples of mobile computing systems are laptops, PDA’s and mobile phones.
There are two distinct types of mobility.
(i) User Mobility
(ii) Device portability.

SIA Group
1.10 Cloud Computing [JNTU-hyderabad]
1. User Mobility
In user mobility, the user is mobile and services follow him. i.e., user can point to same or identical telecommunication
services at distinct places. Techniques that sustain user mobility are call forwarding facility from telephone or desktops supporting
roaming.
2. Device Portability
In device portability, the communication device moves with or without the help of user. Several techniques in the network
guarantees that communication is even possible while the device is being moved. Example for device portability is mobile phones,
where the system automatically directs the device from one base station to the succeeding one, if the signal turns into weak signal.
The mobile computing supports wireless media communication where in devices are small mostly handheld devices. On
the other hand, the mobile communication concerning voice applications such as cellular phones has widely set platform across
the world. It has a huge growth in all aspects and holds large number of subscribers in various cellular networks. When this
technology is upgraded it gives rise to another level where in small devices such as smartphones are implemented to transmit and
receive data across multiple cellular networks. Subsequently, the technology substantiates various applications such as video call
or conferencing.
Therefore, mobile computing-based applications exhibits various upgraded technologies.
Q17. Write short notes on,
(i) Quantum computing
(ii) Optical computing
(iii) Nano computing.
Answer : Model Paper-III, Q3

(i) Quantum Computing

The quantum computing came into existence inorder to solve the problem of manufacturers of computing systems. The
problem is embedding of more and more transistors into tiny spaces of integrated circuits. This addition demands equivalent
processing power i.e., the processing power should be doubled in every 18 months. So, the new quantum computing based solution
comes into picture where the entire focus is made on quantum information.
Typically, the quantum computers are way faster in opposed to most powerful super computers in present. The functionality
of these quantum computing is different at basic level in comparison to the present technology. They have some specific set of
rules (prototypes). Although, they serve in solving many computing problems but couldn’t replace the present day silicon-based
machines.
(ii) Optical Computing
The Optical computing carries out digital computation through photons in visible light or infrared beams in oppose to
electric current. This is due to the fact that, electric current propagates at about 10% of light speed. Thus, confining the flow of
data rate across long distance. As a consequence, optical fiber came into existence.
The benefits of visible and IR networks can be applied at the device and component scale inorder to develop a computer
which can carryout operations 10 times faster than the traditional computer systems.
(iii) Nano Computing
The nano computing can be defined as a computing systems developed using nanoscale components. Here, the transistors
are used based on carbon nanotubes rather than silicon transistors.
The nanocomputers includes the scale and integration of nanotubes or components. In doing so, the problems of scale can
be concerned with the dimensions of components which comes up to the few nanometer in two dimensions. On the other hand,
the problems concerned with the integration of components include twofold. Here, at first the developer of complex arbitrary
patterns can stay economically infeasible. And secondly, the nanocomputer holds huge quantities of devices which eventually
brings problems. However, this field of study is still under research.

L o o k f o r t h e SIA Group logo o n t h e title cover before you buy