UNIT-1

1)Parallel and Distributed systems architecture and communication

protocols?
A)

What is Parallel Computing?

Parallel computing is a computing approach that involves the simultaneous execution of multiple
tasks or instructions on multiple processing units, such as CPUs or GPUs, to solve a problem faster.
Parallel computing can be applied at different levels, including instruction-level parallelism, thread-
level parallelism, and task-level parallelism.

Parallel computing can significantly improve the performance and efficiency of computation-
intensive tasks, such as scientific simulations, data analytics, and machine learning. It allows for the
efficient utilization of resources by dividing the tasks into smaller chunks that can be processed in
parallel, resulting in faster results and improved productivity.

Parallel computer architecture:

Our discussion of parallel computer architectures starts with the recognition that parallelism at
different levels can be exploited. These levels are:
• Bit-level parallelism
The number of bits processed per clock cycle, often called a word size, has increased gradually from
4-bit processors to 8-bit, 16-bit, 32-bit, and, since 2004, 64-bit.

This has reduced the number of instructions required to process larger operands and allowed a
significant performance improvement.

During this evolutionary process the number of address bits has also increased, allowing instructions
to reference a larger address space.

• Instruction-level parallelism
Today’s computers use multi-stage processing pipelines to speed up execution. Once an n-stage
pipeline is full, an instruction is completed at every clock cycle.

A “classic” pipeline of a Reduced Instruction Set Computing (RISC) architecture consists of five stages:
instruction fetch, instruction decode, instruction execution, memory access, and write back. A
Complex Instruction Set Computing (CISC) architecture could have a much large number of pipeline
stages; for example, an Intel Pentium 4 processor has a 35-stage pipeline.

• Data parallelism or loop parallelism

The program loops can be processed in parallel.

• Task parallelism
The problem can be decomposed into tasks that can be carried out concurrently. A widely used type
of task parallelism is the Same Program Multiple Data (SPMD) paradigm.

As the name suggests, individual processors direct the same program but on different portions of the
input data. Data dependencies cause different flows of control in individual processors.

In 1966 Michael Flynn proposed a classification of computer architectures based on the number of
concurrent control/instruction and data streams:

 (SISD) Single Instruction, Single Data

 (SIMD) Single Instruction, Multiple Data

 (MISD) Multiple Instruction, Single Data

 (MIMD) Multiple Instruction, Multiple Data

The SIMD architecture supports vector processing. When an SIMD instruction is issued, operations
on individual vector components are carried out concurrently.

The desire to support real-time graphics with vectors of two, three, or more dimensions led to the
development of graphic processing units (GPUs). GPUs are very efficient at the parallel processing of
large blocks of data, and their highly parallel structures based on SIMD execution are used in
embedded systems, mobile phones, personal computers, workstations, and game consoles. Nvidia
and AMD/ATI produce most GPUs.

The MIMD architecture refers to a system with several processors that function asynchronously and
independently. At any point in time, different processors may be executing different instructions on
different pieces of data.
MIMD and SPMD can be further classified into shared memory (Uniform Memory Access – UMA,
Non-uniform Memory Access – NUMA) and distributed memory (Cache-Only Memory Access –
COMA) systems.

Benefits of Parallel Computing:

l) Faster Processing: Parallel computing allows for concurrent execution of tasks, leading to faster
results compared to sequential processing.

2) Improved Scalability: Parallel computing enables the use of multiple processing units, making it
easier to scale up the computation as needed.

3) Better Resource Utilization: Parallel computing efficiently utilizes the available resources, such as
CPU cores or GPUs, leading to improved resource utilization and cost savings.

4) Enhanced Problem Solving: Parallel computing can handle complex computations and large
datasets, enabling faster and more accurate solutions to complex problems.

5) Future-proofing: As technology advances, parallel computing becomes increasingly relevant, as it

allows for better utilization of modern hardware, such as multi-core processors and GPUs.

Some examples of real-world applications of parallel computing include:

l )High-performance computing clusters used for scientific research, such as NASA’s Pleiades cluster.

2) Graphics processing units (GPUs) used for machine learning and deep learning.

3) Distributed computing frameworks like Apache Spark and Apache Flink.

What are Distributed Systems?

A distributed system is a group of independent computers or servers that work together as a single
system, sharing resources, data, and computations across multiple nodes. Each node in a distributed
system communicates and collaborates with other nodes to achieve a common goal. The nodes can
be located in the same place (e.g., data center) or spread across different locations.

Distributed Systems:
A distributed system is a collection of autonomous computers that are connected through a network
and use a distribution software called middleware, which enables computers to coordinate their
activities and share the resources of the system. A distributed system’s users perceive the system as a
single computing facility.

A distributed system has the following characteristics:

 Its components are autonomous.

 Scheduling and other resource management are decentralized.

 There are multiple points of control (security policies are implemented by each system).

 There are multiple points of failure, and the resources may not be accessible at all times.

Distributed systems can be scaled by adding additional resources and can be designed to maintain
availability even at low levels of hardware/software/network reliability. Distributed systems have
been around for several decades.

Modern operating systems allow a user to mount a remote file system and access it the same way a
local file system is accessed, yet with a performance penalty due to larger communication costs.

The Remote Procedure Call (RPC) supports inter-process communication and allows a procedure on a
system to invoke a procedure running in a different address space, possibly on a remote system. RPCs
were introduced in the early 1980s by Bruce Nelson and used for the first time at Xerox. The Network
File System (NFS) introduced in 1984 was based on Sun’s RPC.

Many programming languages support RPCs; for example, Java Remote Method Invocation (Java
RMI) provides functionality similar to that of UNIX RPC methods, and XML-RPC uses XML to encode
HTTP-based calls.

Middleware should support a set of desirable properties of a distributed system:

 Access transparency: Local and remote information objects are accessed using identical
operations.

 Location transparency: Information objects are accessed without knowledge of their

location.

 Concurrency transparency: Several processes run concurrently using shared information

objects without interference.

 Replication transparency: Multiple instances of information objects are used to increase

reliability without affecting applications.

 Failure transparency: The concealment of faults; migration transparency ensures

information objects are moved without affecting operations.

 Performance transparency: The system can be reconfigured based on performance or

quality of service requirements.

 Scaling transparency: The system and the applications can scale without a change in system
structure and without affecting operations.

Benefits of Distributed Systems:

l )Scalability: Distributed systems can handle a large amount of data
and tasks by adding more nodes to the system as needed, making
them highly scalable.
2) Fault Tolerance: The fault tolerance feature is an inherent design
principle of distributed systems that enables them to withstand and
recover from failures. If a node fails, the workload can be
automatically shifted to other available nodes, ensuring high
availability and reliability.
3)High Performance: Parallel processing in distributed systems
allows for faster execution of tasks by dividing them among multiple
nodes, leading to improved performance.
4) Flexibility: Distributed systems allow for flexibility in adding or
removing nodes based on the changing requirements of the system.
5) Cost-Effectiveness: Distributed systems can be cost-effective, as
they can utilize commodity hardware and distribute the workload
across multiple nodes, reducing the need for expensive hardware.
Some examples of real-world applications of distributed
systems include:
l) Companies such as Amazon, Google, and Microsoft offer cloud
computing services through their respective platforms, namely
Amazon Web Services, Google Cloud Platform, and Microsoft Azure
2) Distributed databases like Cassandra, MongoDB, and Apache
HBase.
3) Hadoop Distributed File System (HDFS) exemplifies a type of file
system that distributes data across multiple computing nodes in a
cluster.

communication protocols :
Communication protocols are vital in distributed systems for enabling
reliable and efficient interaction between nodes. This article delves
into the types, significance, and specific protocols used to manage
communication in distributed environments, ensuring data
consistency and system functionality.

Key Communication Protocols in Distributed Systems

In distributed systems, several key communication protocols are
commonly used to facilitate interaction between nodes, manage data
exchange, and ensure system consistency. Here’s an overview of some
of the most important communication protocols:

1. Message Passing Protocols

Purpose: Facilitate communication by sending messages between
nodes. Messages can be sent synchronously or asynchronously,
depending on the protocol and use case.
How It Works: Nodes send messages that can include commands,
queries, or data. These messages are managed and routed by the
protocol to ensure correct delivery.
Examples:
Message Queuing Telemetry Transport (MQTT): Lightweight,
publish-subscribe messaging protocol ideal for low-bandwidth, high-
latency networks.
Advanced Message Queuing Protocol (AMQP): A protocol
designed for business messaging, providing robust queuing, routing,
and delivery capabilities.
ZeroMQ: A high-performance messaging library that supports
multiple messaging patterns like publish-subscribe, request-reply, and
more.
2 . Object Request Brokers (ORBs)
Purpose: Manage communication between distributed objects,
allowing objects to interact as if they were on the same machine,
despite being on different machines.
How It Works: ORBs handle requests from clients and route them to
the appropriate server-side objects. They abstract the details of
network communication from the objects themselves.
Examples:
Common Object Request Broker Architecture (CORBA): A
standard for object communication that supports multiple
programming languages and platforms.
5. Service-Oriented Architecture (SOA) Protocols
Purpose: Facilitate communication between services in a service-
oriented architecture, often using standardized message formats.
How It Works: Services expose their functionality through well-defined
interfaces, and communication between services is handled through
these interfaces.
Examples:
Simple Object Access Protocol (SOAP): A protocol for exchanging
structured information in web services, using XML.
Web Services Choreography Description Language (WS-CDL):
Describes how services interact and collaborate in a choreographed
fashion..
4. Confidentiality
To ensure that data is only accessible to authorized parties and
remains hidden from unauthorized individuals.
Encryption: The primary technique used to achieve confidentiality. It
transforms data into a format that is unreadable without the
appropriate decryption key.
Symmetric Encryption: Uses a single key for both encryption and
decryption (e.g., AES, DES). It is fast but requires secure key
distribution.
Asymmetric Encryption: Uses a pair of keys (public and private) for
encryption and decryption (e.g., RSA, ECC). It simplifies key
management but is generally slower than symmetric encryption.
5. Integrity
To ensure that data is not altered or tampered with during
transmission.
Hash Functions: Create a unique hash value for the data. Any
change in the data will result in a different hash value, allowing the
detection of tampering.
MD5: Produces a 128-bit hash value, though it's considered weak
against collisions.
SHA-2: Includes a family of hash functions (e.g., SHA-256) that are
more secure than MD5.
Message Authentication Codes (MACs): Combine a secret key with
the message to create a code that verifies both the data’s integrity and
authenticity (e.g., HMAC with SHA-256).
6 . Authentication
To verify the identity of communicating parties and ensure that they
are who they claim to be.
Digital Signatures: Use asymmetric encryption to create a signature
based on the data and a private key. The recipient can verify the
signature using the sender’s public key, ensuring authenticity and
integrity.
RSA, DSA, ECDSA: Common algorithms for generating and verifying
digital signatures.
Certificates: Digital certificates, issued by Certificate Authorities
(CAs), contain public keys and authentication information. They help
in verifying the identity of parties in communication.
X.509 Certificates: Standard format for digital certificates used in
SSL/TLS.
7. Non-Repudiation
 To ensure that a party cannot deny having participated in a
communication or transaction.
Digital Signatures: Provide non-repudiation by proving that the
signature was created by a specific entity and that the entity cannot
deny its involvement.
8. Authorization
To control access to resources and ensure that users have permission
to perform specific actions.
Access Control Lists (ACLs): Define which users or systems are
permitted to access certain resources or perform actions.
Role-Based Access Control (RBAC): Assigns permissions based on
roles within an organization, simplifying management of user rights.

2)cloud computing delivery models(Iaas,saas,paas)

Cloud Computing means using the internet to store, manage, and process data instead of using your
own computer or local server. The data is stored on remote servers, that are owned by companies
called cloud providers such as Amazon, Google, Microsoft). These companies charge you based on
how much you have used their services.

Types of Cloud Computing

Most cloud computing services fall into five broad categories:

1. Software as a service (SaaS)

2. Platform as a service (PaaS)

 3. Infrastructure as a service (IaaS)

4. Anything/Everything as a service (XaaS)

5. Function as a Service (FaaS)

These are sometimes called the cloud computing stack because they are built on top of one another.
Knowing what they are and how they are different, makes it easier to accomplish your goals. These
abstraction layers can also be viewed as a layered architecture where services of a higher layer can
be composed of services of the underlying layer i.e, SaaS can provide Infrastructure.

1. Software as a
Service(SaaS)

Software-as-a-Service (SaaS) means using software over the internet instead of installing in on your
computer. You don't have to worry about downloading, updating, or maintaining anything- the
company that provides the software handles all of that.

Example:

Think of Google Docs. You don't need to install it. You just open your browser, log in, and start using
it. Google stores your work and keeps the software updated. You just use it when you need it.

SaaS is usually offered on pay-as-you-go basis, and you can access it from any device with internet.
It's also called web-based-software or on-demand software because you can use it anytime,
anywhere, without setup.

Advantages of SaaS

Cost-Effective: Pay only for what you use.

Reduced time: Users can run most SaaS apps directly from their web browser without needing to
download and install any software. This reduces the time spent in installation and configuration and
can reduce the issues that can get in the way of the software deployment.

Accessibility: We can Access app data from anywhere.

Automatic updates: Rather than purchasing new software, customers rely on a SaaS provider to
automatically perform the updates.

Scalability: It allows the users to access the services and features on-demand.

The various companies providing Software as a service are Cloud9 Analytics, Salesforce.com, Cloud
Switch, Microsoft Office 365, Big Commerce, Eloqua, dropBox, and Cloud Tran.
Disadvantages of Saas :

Limited customization: SaaS solutions are typically not as customizable as on-premises software,
meaning that users may have to work within the constraints of the SaaS provider's platform and may
not be able to tailor the software to their specific needs.

Dependence on internet connectivity: SaaS solutions are typically cloud-based, which means that
they require a stable internet connection to function properly. This can be problematic for users in
areas with poor connectivity or for those who need to access the software in offline environments.

Security concerns: SaaS providers are responsible for maintaining the security of the data stored on
their servers, but there is still a risk of data breaches or other security incidents.

Limited control over data: SaaS providers may have access to a user's data, which can be a concern
for organizations that need to maintain strict control over their data for regulatory or other reasons.

2. Platform as a Service

PaaS is a type of cloud service that gives developers the tools they need to build and launch apps
online without setting up any hardware and software themselves.

With PaaS, everything runs on the provider's server and is accessed through a web browser. The
provider takes care of things like servers, storage, and operating systems. Developers just focus on
writing and managing the app.

Example:

Imagine you're planning a school's annual day event. You have two options.

Build the Venue yourself (buy land, set up a stage, arrange lighting, etc.).

Or rent a ready-to-use venue and just focus on the actual event.

PaaS is like renting the venue- it saves time, efforts, and setup costs, so you can completely focus on
what matters: building your app.

You don't control the back-end (like servers), but you do control the app you create and how it
behaves.

Advantages of PaaS:

Simple and convenient for users: It provides much of the infrastructure and other IT services, which
users can access anywhere via a web browser.

Cost-Effective: It charges for the services provided on a per-use basis thus eliminating the expenses
one may have for on-premises hardware and software.

Efficiently managing the lifecycle: It is designed to support the complete web application lifecycle:
building, testing, deploying, managing, and updating.

Efficiency: It allows for higher-level programming with reduced complexity thus, the overall
development of the application can be more effective.

The various companies providing Platform as a service are Amazon Web services Elastic Beanstalk,
Salesforce, Windows Azure, Google App Engine, cloud Bees and IBM smart cloud.

Disadvantages of Paas:
Limited control over infrastructure: PaaS providers typically manage the underlying infrastructure
and take care of maintenance and updates, but this can also mean that users have less control over
the environment and may not be able to make certain customizations.

Dependence on the provider: Users are dependent on the PaaS provider for the availability,
scalability, and reliability of the platform, which can be a risk if the provider experiences outages or
other issues.

Limited flexibility: PaaS solutions may not be able to accommodate certain types of workloads or
applications, which can limit the value of the solution for certain organizations.

3. Infrastructure as a Service

Infrastructure as a service (IaaS) is a cloud service where companies rent IT resources like servers,
storage, and networks instead of buying and managing them.

It's like outsourcing your computer hardware. The cloud provider gives you the basic building blocks
(like virtual machines, storage, and internet access), and you use them to run your apps and services.

You pay based on how much you use - by the hour, week, or month. The way, you don't need to
spend a lot of money on buying hardware.

Example:

Imagine you want to start a website. Instead of buying you own server, you rent on a cloud provider's
server. You use their storage and networking, but you control what runs on it-like your website or
app.

That's IaaS, You get the flexibility and power of your own setup, without the cost and trouble of
maintaining hardware.

Advantages of IaaS:

Cost-Effective: Eliminates capital expense and reduces ongoing cost and IaaS customers pay on a per-
user basis, typically by the hour, week, or month.

Website hosting: Running websites using IaaS can be less expensive than traditional web hosting.

Security: The IaaS Cloud Provider may provide better security than your existing software.

Maintenance: There is no need to manage the underlying data center or the introduction of new
releases of the development or underlying software. This is all handled by the IaaS Cloud Provider.

The various companies providing Infrastructure as a service are Amazon web services, Bluestack,
IBM, Openstack, Rackspace, and Vmware.

Disadvantages of laaS :

Limited control over infrastructure: IaaS providers typically manage the underlying infrastructure
and take care of maintenance and updates, but this can also mean that users have less control over
the environment and may not be able to make certain customizations.

Security concerns: Users are responsible for securing their own data and applications, which can be a
significant undertaking.
Limited access: Cloud computing may not be accessible in certain regions and countries due to legal
policies.

4. Anything as a Service

It is also known as Everything as a Service. Most of the cloud service providers nowadays offer
anything as a service that is a compilation of all of the above services including some additional
services.

Advantages of XaaS:

Scalability: XaaS solutions can be easily scaled up or down to meet the changing needs of an
organization.

Flexibility: XaaS solutions can be used to provide a wide range of services, such as storage,
databases, networking, and software, which can be customized to meet the specific needs of an
organization.

Cost-effectiveness: XaaS solutions can be more cost-effective than traditional on-premises solutions,
as organizations only pay for the services.

Disadvantages of XaaS:

Dependence on the provider: Users are dependent on the XaaS provider for the availability,
scalability, and reliability of the service, which can be a risk if the provider experiences outages or
other issues.

Limited flexibility: XaaS solutions may not be able to accommodate certain types of workloads or
applications, which can limit the value of the solution for certain organizations.

Limited integration: XaaS solutions may not be able to integrate with existing systems and data
sources, which can limit the value of the solution for certain organizations.

5. Function as a Service

FaaS is a cloud service that lets you run small pieces of code - called functions- without managing any
servers. You just write your code, upload it, and it runs only when triggered by an event, like a
button click or a file upload.

FaaS is event-driven, meaning the code runs only when something specific happens. You don't need
to keep a sever running in the background - it starts automatically when needed and stops when the
job is done. That's why it is also called serverless (even though servers are still used, they' re
managed entirely by the provider).

Example:

Imagine an online photo app that resizes images whenever a user uploads a photo. With FaaS, you
write a small functions to resize the image. The Function only runs when a photo is uploaded- and
you only pay for the execution.

Advantages of FaaS

Highly Scalable: Auto scaling is done by the provider depending upon the demand.

Cost-Effective: Pay only for the number of events executed.

Code Simplification: FaaS allows the users to upload the entire application all at once. It allows you
to write code for independent functions or similar to those functions.

Maintenance of code is enough and no need to worry about the servers.

The various companies providing Function as a Service are Amazon Web Services - Firecracker,
Google - Kubernetes, Oracle - Fn, Apache OpenWhisk - IBM, OpenFaaS,

Disadvantages of FaaS

Cold start latency: Since FaaS functions are event-triggered, the first request to a new function may
experience increased latency as the function container is created and initialized.

Limited control over infrastructure: FaaS providers typically manage the underlying infrastructure
and take care of maintenance and updates, but this can also mean that users have less control over
the environment and may not be able to make certain customizations.

Security concerns: Users are responsible for securing their own data and applications, which can be a
significant undertaking.

Limited scalability: FaaS functions may not be able to handle high traffic or large number of
requests.

3)Advantages of cloud over traditional distributed

systems
Aspect Cloud Computing Traditional Distributed Systems

Requires high capital expenditure for

Pay-as-you-go model; no large upfront
Cost Efficiency infrastructure purchase and
investment for hardware/software.
maintenance.

Scaling requires purchasing and

Easily scalable up or down based on
Scalability configuring new hardware, which is
demand with minimal effort.
time-consuming.

Managed by cloud service provider Maintenance is manual and handled by

Maintenance
(automatic updates, monitoring). in-house IT teams.

Usually limited to specific network or

Accessibility Accessible from anywhere via the internet.
locations.

Resource Virtualization allows optimal resource Often underutilized due to static

Utilization usage and sharing. allocation of resources.

Disaster Built-in backup and disaster recovery Requires separate backup systems and
Recovery options. recovery planning.

Deployment Rapid deployment of applications and Deployment takes longer due to

Speed services. hardware/software setup.

Flexibility Supports multiple platforms, services, and Limited flexibility; integrating new tech
integration with emerging technologies like
Aspect Cloud Computing Traditional Distributed Systems

AI & IoT. requires manual adaptation.

Energy Optimized data centers with energy- Often less energy-efficient due to
Efficiency efficient designs. smaller, less optimized setups.

Security Providers ensure regular security patches Security depends on local

Updates and compliance. administration, often slower to update.

4)How cloud computing addressed some of

limitations of former technologies which have offered
similar services?
Limitation in Former Technologies Cloud Computing Solution

1. High upfront hardware and software costs — Pay-as-you-go model – Users rent computing
Traditional systems required organizations to buy power, storage, and applications only for the time
and maintain their own servers, storage, and they use them, avoiding large capital
licenses. expenditure.

On-demand scalability – Resources (CPU, RAM,

2. Limited scalability — Adding resources in
storage) can be scaled up or down instantly
clusters or grids was slow, required downtime, and
through virtualization and automation without
was often hardware-dependent.
downtime.

3. Complex setup and configuration —

Simplified provisioning – Cloud platforms
Distributed and grid systems often needed
automate deployment via management consoles,
specialized expertise to configure hardware,
APIs, and Infrastructure-as-Code tools.
middleware, and load balancing.

Virtualization and containerization – Enable

4. Low flexibility — Older systems were usually
running multiple isolated environments on the
tied to specific physical infrastructure and could
same hardware, easily reallocated for different
not adapt easily to changing workloads.
workloads.

5. Maintenance overhead — Organizations had to Managed services – Cloud providers handle

manage updates, patches, backups, and hardware maintenance, updates, security patches, and
replacement manually. backups automatically.

Global access via internet – Cloud services are

6. Limited geographic accessibility — Resources
accessible from anywhere with proper
were tied to physical locations, making global
authentication, enabling remote work and global
collaboration difficult.
collaboration.

7. Poor fault tolerance — Hardware failure in Built-in redundancy and disaster recovery –
traditional setups could cause long downtimes Cloud providers replicate data across multiple
without expensive redundancy. locations with automatic failover.

8. Inefficient resource utilization — Servers in Resource pooling – Virtualization allows shared

Limitation in Former Technologies Cloud Computing Solution

older systems were often underused because they use of physical resources across many users,
were dedicated to specific tasks. improving utilization and efficiency.

9. Slow service deployment — Launching new Rapid provisioning – New virtual machines,
services required purchasing hardware, setting up databases, or applications can be launched in
environments, and manual configuration. minutes using cloud automation.

Service models (IaaS, PaaS, SaaS) – Cloud

10. Limited service variety — Grid/cluster systems
provides complete solutions: infrastructure,
mainly provided raw computation power, not end-
development platforms, and ready-to-use
user services.
applications.

5)peer to peer systems

peer-to –peer systems in cloud computing

What is P2P (Peer-to-Peer Process)?

The P2P process deals with a network structure where any participant in the network known as a
node acts as both a client and a server. This means that, rather than relying on a basis server to
supply resources or services, everybody from the network of nodes can trade resources and services
with one another. In a P2P system, every node has an equal role to play and the same functionalities,
which means that the loads are well shared.

What Is a Peer-to-Peer (P2P) Service?

A peer-to-peer network is a simple network of computers. It first came into existence in the late
1970s. Here each computer acts as a node for file sharing within the formed network. Here each
node acts as a server and thus there is no central server in the network. This allows the sharing of a
huge amount of data. The tasks are equally divided amongst the nodes. Each node connected in the
network shares an equal workload. For the network to stop working, all the nodes need to
individually stop working. This is because each node works independently.

Types of P2P Networks

 Unstructured P2P Networks: In this type of P2P network, each device is able to make an
equal contribution. This network is easy to build as devices can be connected randomly in
the network. But being unstructured, it becomes difficult to find content. For example,
Napster, Gnutella, etc.

 Structured P2P Networks: It is designed using software that creates a virtual layer in order to
put the nodes in a specific structure. These are not easy to set up but can give easy access to
users to the content. For example, P-Grid, Kademlia, etc.

 Hybrid P2P Networks: It combines the features of both P2P networks and client-server
architecture. An example of such a network is to find a node using the central server.
P2P Network Architecture

In the P2P network architecture, the computers connect with each other in a workgroup to share
files, and access to internet and printers.

Each computer in the network has the same set of responsibilities and capabilities.

Each device in the network serves as both a client and server.

The architecture is useful in residential areas, small offices, or small companies where each computer
act as an independent workstation and stores the data on its hard drive.

Each computer in the network has the ability to share data with other computers in the network.

The architecture is usually composed of workgroups of 12 or more computers.

Applications of P2P Network

Below are some of the common uses of P2P network:

File Sharing: P2P network is the most convenient, cost-efficient method for file sharing for
businesses. Using this type of network there is no need for intermediate servers to transfer the file.

Blockchain: The P2P architecture is based on the concept of decentralization. When a peer-to-peer
network is enabled on the blockchain it helps in the maintenance of a complete replica of the
records ensuring the accuracy of the data at the same time. At the same time, peer-to-peer networks
ensure security also.

Direct Messaging: P2P network provides a secure, quick, and efficient way to communicate. This is
possible due to the use of encryption at both the peers and access to easy messaging tools.
Collaboration: The easy file sharing also helps to build collaboration among other peers in the
network.

File Sharing Networks: Many P2P file sharing networks like G2, and eDonkey have popularized peer-
to-peer technologies.

Content Distribution: In a P2P network, unline the client-server system so the clients can both
provide and use resources. Thus, the content serving capacity of the P2P networks can actually
increase as more users begin to access the content.

IP Telephony: Skype is one good example of a P2P application in VoIP.

Examples of P2P Networks

P2P networks can be basically categorized into three levels.

The first level is the basic level which uses a USB to create a P2P network between two systems.

The second is the intermediate level which involves the usage of copper wires in order to connect
more than two systems.

The third is the advanced level which uses software to establish protocols in order to manage
numerous devices across the internet.

Some of the popular P2P networks are Gnutella, BitTorrent, eDonkey, Kazaa, Napster, and Skype.

Advantages of P2P Network

Easy to Maintain: The network is easy to maintain because each node is independent of the other.

Less Costly: Since each node acts as a server, therefore the cost of the central server is saved. Thus,
there is no need to buy an expensive server.

No Network Manager: In a P2P network since each node manages his or her own computer, thus
there is no need for a network manager.

Adding Nodes is Easy: Adding, deleting, and repairing nodes in this network is easy.

Less Network Traffic: In a P2P network, there is less network traffic than in a client/ server network.

Disadvantages of P2P Network

Data is Vulnerable: Because of no central server, data is always vulnerable to getting lost because of
no backup.

Less Secure: It becomes difficult to secure the complete network because each node is independent.

Slow Performance: In a P2P network, each computer is accessed by other computers in the network
which slows down the performance of the user.

Files Hard to Locate: In a P2P network, the files are not centrally stored, rather they are stored on
individual computers which makes it difficult to locate the files.

6)Ethical issues and vulnerability

Ethical Issues in Cloud Computing
Ethical issues in cloud computing revolve around data privacy, security, and ownership concerns.
Storing confidential information on third-party servers raises questions about who has access to the
data and how it is being used. Additionally, the potential for data breaches and unauthorized access
can compromise the trust between users and cloud service providers. It is important to establish
clear policies and regulations to address these ethical concerns and ensure the protection of
sensitive information in the cloud.

1) Data Privacy: When offering a training program in cloud computing to students, there is a risk of
compromising the privacy of their personal information stored in the cloud. It is important to ensure
that proper security measures are in place to protect sensitive data.

2) Security Concerns: Cloud computing systems are vulnerable to security breaches, hacking
attempts, and unauthorized access. Training programs should cover best practices for securing data
and preventing potential cyber threats.

3) Compliance with Regulations: There are various regulations and laws governing the storage and
processing of data, such as GDPR and HIPAA. It is essential to educate students about compliance
requirements when using cloud services.

4) Vendor Lock In: Students should be aware of the risks associated with becoming dependent on a
specific cloud service provider, which could limit their ability to switch providers in the future.

5) Data Ownership: Clarifying data ownership rights is crucial in cloud computing. Students should
understand who owns the data they store in the cloud and what happens to their data if they
terminate their relationship with a cloud provider.

6) Accessibility and Availability: Cloud services are prone to outages and downtime, affecting the
availability of data and applications. Training programs should address strategies for ensuring
continuity and minimizing disruptions.

7) Environmental Impact: Cloud computing has a carbon footprint due to the energy consumption of
data centers. Educating students about the environmental impact of cloud services can raise
awareness and promote sustainable practices.

8) Intellectual Property Rights: Issues related to intellectual property rights can arise when students
share or collaborate on cloud based resources. It is essential to address copyright concerns and
proper usage of intellectual property in training programs.

9) Transparency and Accountability: Cloud providers should be transparent about their data
practices and security measures. Students should be educated on how to hold providers accountable
for protecting their data and upholding ethical standards.

10) Ethical Use of Data: Students should be trained on the ethical implications of collecting,
analyzing, and sharing data in the cloud. Emphasizing responsible data practices can help prevent
misuse and unethical behavior.

11) Bias in Algorithms: Cloud computing often involves the use of algorithms and AI technologies,
which can perpetuate biases and discriminatory outcomes. Training programs should discuss the
importance of fairness and equity in algorithm design and implementation.

12) Cultural Sensitivity: When offering training programs to a diverse group of students, considering
cultural differences and sensitivities is paramount. Cloud computing education should be inclusive
and respectful of varying perspectives.
13) Social Responsibility: It is essential for students to understand the social impact of cloud
computing, such as its role in promoting digital inclusion or exacerbating digital divides. Encouraging
ethical considerations in using cloud services can contribute to positive societal outcomes.

14) Trust and Reliability: Building trust in cloud services is crucial for user adoption and acceptance.
Training programs should emphasize the importance of reliability, transparency, and open
communication in fostering trust between users and cloud providers.

15) Continuous Learning and Improvement: Finally, ethical issues in cloud computing are constantly
evolving, requiring students to engage in continuous learning and self improvement. Encouraging a
culture of ethical reflection and adaptation can help students navigate ethical dilemmas in the
dynamic field of cloud computing.

vulnerabilities in cloud computing

Cloud vulnerabilities are weaknesses in your cloud environment that can be exploited by attackers.
These vulnerabilities can lead to unauthorised access, data theft, or service disruptions. Despite their
commonality, many organisations fail to address these security gaps, leaving them exposed to
potential breaches.

A cloud vulnerability is different from a threat. While a threat is an immediate danger, like a
cyberattack, a vulnerability is the weakness that allows the threat to cause harm. For example, poor
access management is a vulnerability that could give attackers access to sensitive data.

The top cloud vulnerabilities

Cloud vulnerabilities can be found across your cloud infrastructure, applications, and storage. Some
of the top cloud vulnerabilities include:

Misconfigurations

This is one of the most common cloud vulnerabilities. Misconfigurations occur when there are errors
in the security settings of cloud applications and systems. These can happen in virtual machines,
containers, and other cloud infrastructures due to administrative mistakes or a lack of awareness.
Misconfigurations are a leading cause of data breaches, often resulting from open ports,
overprivileged user accounts, unsecured storage (like open S3 buckets), and using default passwords.

Lack of visibility

Many enterprises use a mix of cloud technologies from various providers, creating complex IT
environments. This can lead to scattered vulnerabilities that are difficult to identify. Without a clear
view of the entire cloud ecosystem, it is challenging to assess and manage risks effectively.

Poor access management

In cloud environments, digital identities often outnumber human identities. This makes Identity
Access Management (IAM) critical, as poor access management can serve as a gateway for
cybercriminals. Vulnerabilities may include weak password practices, a lack of Multi-Factor
Authentication (MFA), and over-privileged accounts.

Insider threats
Insider threats come from individuals who already have access to the organisation’s IT environment,
including employees and third-party vendors. Cloud vulnerabilities in cloud computing can arise
from accidental errors or intentional actions, and they are often more damaging due to the insider's
knowledge of the system.

Unsecured APIs

Cloud Application Programming Interfaces (APIs) enable communication between cloud applications.
However, unsecured APIs can expose your organisation to significant risks, including weak
authentication and incorrect access controls. These vulnerabilities can allow unauthorised access to
sensitive data.

Zero-days

Zero-day vulnerabilities refer to flaws that are unknown to the vendor, allowing cybercriminals to
exploit them before a fix is available. These can lead to serious attacks on software and systems,
especially if not monitored regularly.

Shadow IT

Shadow IT occurs when employees use cloud services without official approval from the IT
department. This can create cloud security risks related to data loss and unauthorised access, as
unmonitored services may not meet the organisation’s security standards.

Lack of encryption

When sensitive data is not encrypted, unauthorised individuals can easily access it if they breach
your cloud environment. Encryption protects data by transforming it into unreadable formats unless
you have the proper keys. Without it, the risk of data exposure, one of the top cloud vulnerabilities,
increases significantly.

How to overcome cloud vulnerabilities

For mitigating cloud vulnerabilities, organisations need a strong cybersecurity strategy that
addresses potential threats and safeguards data effectively. Here are key actions you can take to
strengthen your cloud security and prevent breaches:

1. Create strong and regularly updated passwords

Creating secure passwords is one of the easiest and most effective strategies to avoid cloud
vulnerabilities. Make sure they include a variety of uppercase and lowercase letters, numerals, and
special characters. Additionally, it's crucial to change passwords regularly to reduce the risk of
unauthorised access.

2. Implement security awareness programs

Your workers, customers, and other stakeholders should be informed on cloud vulnerabilities in
cloud computing. Conduct security awareness workshops to teach everyone how to recognise
dangers such as phishing (fake emails or texts that mislead you into disclosing important information)
and how to manage data securely. Awareness at every level minimises the chance of human errors
leading to breaches.

3. Seek expert consultation

It can be helpful to bring in cybersecurity experts to assess your cloud security. These professionals
can identify risks you may not have noticed and suggest strategies tailored to your specific needs.
Their expertise can help you stay ahead of emerging threats.

4. Conduct frequent data backups

Data loss is a major concern in the cloud. Regular data backups ensure that even in the event of an
attack, such as ransomware (where hackers hold your data hostage for a fee), you can recover your
files and avoid significant disruptions. Ensure your backups are stored securely and are regularly
updated to eradicate the possibility of common cloud vulnerabilities.

5. Use Data Loss Prevention tools

Consider using Data Loss Prevention (DLP) tools to monitor and manage the movement of sensitive
data across your systems. These technologies can alert you to questionable activities, such as illegal
file transfers, so preventing data leaks.

6. Encrypt your data

Encryption is a method of converting data into unreadable formats, ensuring that even if
unauthorised users gain access, they cannot use the information without the decryption key. This is
crucial in mitigating cloud vulnerabilities and preventing attacks like ransomware, where hackers
attempt to hold your data for ransom.

7. Conduct routine security audits

Regular security audits can help you identify weak points in your cloud infrastructure. These audits
involve assessing your systems for vulnerabilities and ensuring that all security measures are up-to-
date and effective in preventing potential attacks.

8. Develop a breach response plan

Prepare for potential breaches by creating a cybersecurity breach response plan. This plan should
outline the steps to take in case of a security incident, including how to contain the breach, who to
notify, and how to recover data quickly. Having a strategy in place allows for a quick and effective
reaction to reduce cloud vulnerabilities.

9. Test your APIs

Since Application Programming Interfaces (APIs) allow different software applications to

communicate with each other, they can be vulnerable to attacks if not properly secured. Conduct API
penetration testing to identify and fix any vulnerabilities in your API security.

10. Regulate user access

Not every employee need access to sensitive data. Implement stringent user access restrictions to
ensure that only permitted persons have access to vital information. This lowers the danger of insider
threats or unintentional data leakage.

11. Limit and revoke access timely

Limiting access to data is essential to reduce the chances of misuse. Regularly review access rights
and revoke access for users who no longer need it. This step is especially important for temporary
employees or third-party vendors.
12. Conduct vulnerability assessments

Regularly assess your systems for weaknesses by conducting cloud vulnerability assessments. This
helps you discover potential flaws in your infrastructure before attackers do and allows you to patch
them in time.

Major challenges

7)Logical Clock in Distributed System

In distributed systems, ensuring synchronized events across multiple nodes is crucial for consistency
and reliability. Enter logical clocks, a fundamental concept that orchestrates event ordering without
relying on physical time. By assigning logical timestamps to events, these clocks enable systems to
reason about causality and sequence events accurately, even across network delays and varied
system clocks. This article explores how logical clocks enhance distributed system design.

Differences Between Physical and Logical Clocks

Physical clocks and logical clocks serve distinct purposes in distributed systems:

Nature of Time:

1.Physical Clocks: These rely on real-world time measurements and are typically synchronized using
protocols like NTP (Network Time Protocol). They provide accurate timestamps but can be affected
by clock drift and network delays.

2. Logical Clocks: These are not tied to real-world time and instead use logical counters or
timestamps to order events based on causality. They are resilient to clock differences between nodes
but may not provide real-time accuracy.`

Usage:

Physical Clocks: Used for tasks requiring real-time synchronization and precise timekeeping, such as
scheduling tasks or logging events with accurate timestamps.

Logical Clocks: Used in distributed systems to order events across different nodes in a consistent and
causal manner, enabling synchronization and coordination without strict real-time requirements.

Dependency:

Physical Clocks: Dependent on accurate timekeeping hardware and synchronization protocols to

maintain consistency across distributed nodes.

Logical Clocks: Dependent on the logic of event ordering and causality, ensuring that events can be
correctly sequenced even when nodes have different physical time readings.
Types of Logical Clocks in Distributed System

1. Lamport Clocks

Lamport clocks provide a simple way to order events in a distributed system. Each node maintains a
counter that increments with each event. When nodes communicate, they update their counters
based on the maximum value seen, ensuring a consistent order of events.

Characteristics of Lamport Clocks:

Simple to implement.

Provides a total order of events but doesn't capture concurrency.

Not suitable for detecting causal relationships between events.

Algorithm of Lamport Clocks:

Initialization: Each node initializes its clock LLL to 0.

Internal Event: When a node performs an internal event, it increments its clock LLL.

Send Message: When a node sends a message, it increments its clock LLL and includes this value in
the message.

Receive Message: When a node receives a message with timestamp T: It sets L=max⁡(L,T)+1

Advantages of Lamport Clocks:

Simple to implement and understand.

Ensures total ordering of events.

2. Vector Clocks

Vector clocks use an array of integers, where each element corresponds to a node in the system.
Each node maintains its own vector clock and updates it by incrementing its own entry and
incorporating values from other nodes during communication.

Characteristics of Vector Clocks:

Captures causality and concurrency between events.

Requires more storage and communication overhead compared to Lamport clocks.

Algorithm of Vector Clocks:

Initialization: Each node PiP_iPi initializes its vector clock ViV_iVi to a vector of zeros.

Internal Event: When a node performs an internal event, it increments its own entry in the vector
clock Vi[i]V_i[i]Vi[i].

Send Message: When a node PiP_iPi sends a message, it includes its vector clock ViV_iVi in the
message.

Receive Message: When a node PiP_iPi receives a message with vector clock Vj:

It updates each entry: Vi[k]=max⁡(Vi[k],Vj[k])

It increments its own entry: Vi[i]=Vi[i]+1

Advantages of Vector Clocks:

Accurately captures causality and concurrency.

Detects concurrent events, which Lamport clocks cannot do.

3. Matrix Clocks

Matrix clocks extend vector clocks by maintaining a matrix where each entry captures the history of
vector clocks. This allows for more detailed tracking of causality relationships.

Characteristics of Matrix Clocks:

More detailed tracking of event dependencies.

Higher storage and communication overhead compared to vector clocks.

Algorithm of Matrix Clocks:

Initialization: Each node PiP_iPi initializes its matrix clock MiM_iMi to a matrix of zeros.

Internal Event: When a node performs an internal event, it increments its own entry in the matrix
clock Mi[i][i]M_i[i][i]Mi[i][i].

Send Message: When a node PiP_iPi sends a message, it includes its matrix clock MiM_iMi in the
message.

Receive Message: When a node PiP_iPi receives a message with matrix clock Mj:

It updates each entry: Mi[k][l]=max⁡(Mi[k][l],Mj[k][l])

It increments its own entry: Mi[i][i]=Mi[i][i]+1

Advantages of Matrix Clocks:

Detailed history tracking of event causality.

Can provide more information about event dependencies than vector clocks.

4. Hybrid Logical Clocks (HLCs)

Hybrid logical clocks combine physical and logical clocks to provide both causality and real-time
properties. They use physical time as a base and incorporate logical increments to maintain event
ordering.

Characteristics of Hybrid Logical Clocks:

Combines real-time accuracy with causality.

More complex to implement compared to pure logical clocks.

Algorithm of Hybrid Logical Clocks:

Initialization: Each node initializes its clock HHH with the current physical time.

Internal Event: When a node performs an internal event, it increments its logical part of the HLC.

Send Message: When a node sends a message, it includes its HLC in the message.
Receive Message: When a node receives a message with HLC T:

It updates its H = max⁡(H,T)+1

Advantages of Hybrid Logical Clocks:

Balances real-time accuracy and causal consistency.

Suitable for systems requiring both properties, such as databases and distributed ledgers.

5. Version Vectors

Version vectors track versions of objects across nodes. Each node maintains a vector of version
numbers for objects it has seen.

Characteristics of Version Vectors:

Tracks versions of objects.

Similar to vector clocks, but specifically for versioning.

Algorithm of Version Vectors:

Initialization: Each node initializes its version vector to zeros.

Update Version: When a node updates an object, it increments the corresponding entry in the
version vector.

Send Version: When a node sends an updated object, it includes its version vector in the message.

Receive Version: When a node receives an object with a version vector:

It updates its version vector to the maximum values seen for each entry.

Advantages of Version Vectors:

Efficient conflict resolution.

Tracks object versions effectively in distributed databases and file systems.

Applications of Logical Clocks

Logical clocks play a crucial role in distributed systems by providing a way to order events and
maintain consistency. Here are some key applications:

Event Ordering

Causal Ordering: Logical clocks help establish a causal relationship between events, ensuring that
messages are processed in the correct order.

Total Ordering: In some systems, it's essential to have a total order of events. Logical clocks can be
used to assign unique timestamps to events, ensuring a consistent order across the system.

Causal Consistency

Consistency Models: In distributed databases and storage systems, logical clocks are used to ensure
causal consistency. They help track dependencies between operations, ensuring that causally related
operations are seen in the same order by all nodes.
Distributed Debugging and Monitoring

Tracing and Logging: Logical clocks can be used to timestamp logs and trace events across different
nodes in a distributed system. This helps in debugging and understanding the sequence of events
leading to an issue.

Performance Monitoring: By using logical clocks, it's possible to monitor the performance of
distributed systems, identifying bottlenecks and delays.

Distributed Snapshots

Checkpointing: Logical clocks are used in algorithms for taking consistent snapshots of the state of a
distributed system, which is essential for fault tolerance and recovery.

Global State Detection: They help detect global states and conditions such as deadlocks or stable
properties in the system.

Concurrency Control

Optimistic Concurrency Control: Logical clocks help detect conflicts in transactions by comparing
timestamps, allowing systems to resolve conflicts and maintain data integrity.

Versioning: In versioned storage systems, logical clocks can be used to maintain different versions of
data, ensuring that updates are applied correctly and consistently.

Challenges and Limitations with Logical Clocks

Logical clocks are essential for maintaining order and consistency in distributed systems, but they
come with their own set of challenges and limitations:

Scalability Issues

Vector Clock Size: In systems using vector clocks, the size of the vector grows with the number of
nodes, leading to increased storage and communication overhead.

Management Complexity: Managing and maintaining logical clocks across a large number of nodes
can be complex and resource-intensive.

Synchronization Overhead

Communication Overhead: Synchronizing logical clocks requires additional messages between

nodes, which can increase network traffic and latency.

Processing Overhead: Updating and maintaining logical clock values can add computational
overhead, impacting the system's overall performance.

Handling Failures and Network Partitions

Clock Inconsistency: In the presence of network partitions or node failures, maintaining consistent
logical clock values can be challenging.

Recovery Complexity: When nodes recover from failures, reconciling logical clock values to ensure
consistency can be complex.

Partial Ordering
Limited Ordering Guarantees: Logical clocks, especially Lamport clocks, only provide partial ordering
of events, which may not be sufficient for all applications requiring a total order.

Conflict Resolution: Resolving conflicts in operations may require additional mechanisms beyond
what logical clocks can provide.

Complexity in Implementation

Algorithm Complexity: Implementing logical clocks, particularly vector and matrix clocks, can be
complex and error-prone, requiring careful design and testing.

Application-Specific Adjustments: Different applications may require customized logical clock

implementations to meet their specific requirements.

Storage Overhead

Vector and Matrix Clocks: These clocks require storing a vector or matrix of timestamps, which can
consume significant memory, especially in systems with many nodes.

Snapshot Storage: For some applications, maintaining snapshots of logical clock values can add to
the storage overhead.

Propagation Delay

Delayed Updates: Updates to logical clock values may not propagate instantly across all nodes,
leading to temporary inconsistencies.

Latency Sensitivity: Applications that are sensitive to latency may be impacted by the delays in
propagating logical clock updates

Message delivery rules in distributed system

In a distributed system, message delivery rules define how messages are sent and received between
different processes. These rules ensure reliable and consistent communication. Common rules
include FIFO (First-In, First-Out) delivery, which guarantees messages are processed in the order they
were sent, and causal ordering, which preserves the order of causally related messages. Other
important aspects include synchronous vs. asynchronous message passing and reliable messaging
which ensures messages are not lost or duplicated

8)concurrency in cloud computing

Concurrency in cloud computing refers to the ability to execute multiple tasks or processes seemingly
simultaneously, improving resource utilization and system responsiveness. It's a fundamental
concept for building scalable and efficient cloud applications. Concurrency is achieved either
through parallel execution on multi-core processors or through time-sharing (context switching) on
single-core systems.

Model concurrency with Petri Nets.

Petri nets are a powerful tool for modeling concurrency in cloud computing. They provide a formal,
graphical, and mathematical framework to represent and analyze concurrent processes, including
those found in distributed systems like cloud environments. By using Petri nets, developers can
model complex interactions, synchronization, and resource sharing within cloud applications,
ensuring proper functionality and performance.

Here's a breakdown of how Petri nets are used in cloud computing:

1. Modeling Concurrent Processes:

 Representing Cloud Components:

Petri nets can represent various cloud components like virtual machines, services, and data storage
as places in the net.

 Modeling Interactions:

Arcs and transitions in the net model interactions between these components, such as data transfer,
service calls, and resource allocation.

 Concurrency and Synchronization:

The graphical nature of Petri nets allows for easy visualization of concurrent activities and
synchronization points, such as resource locking or task scheduling.

2. Analyzing Cloud System Behavior:

 Formal Analysis:

Petri nets offer formal methods for analyzing system behavior, including reachability analysis,
deadlock detection, and performance evaluation.

 Performance Optimization:

By analyzing the Petri net model, developers can identify potential bottlenecks and optimize resource
utilization in the cloud.

 Verification:

Petri nets can be used to verify the correctness of cloud system designs, ensuring that desired
properties like liveness and boundedness are maintained.

3. Applications in Cloud Computing:

 Data Scheduling:

Colored Petri Nets (CPNs) can model data scheduling in cloud storage systems, optimizing data
access and minimizing latency, according to PeerJ.
  Resource Management:

Petri nets can model resource allocation and scheduling in virtualized environments, ensuring
efficient resource utilization and preventing conflicts.

 Cloud File Systems:

Petri nets can be used to model the behavior of cloud file systems, including replication and
pipelining, according to MDPI.

 Cloud-based Applications:

Petri nets can be applied to model and analyze the behavior of various cloud-based applications,
such as e-commerce platforms, social networks, and scientific computing applications.

4. Types of Petri Nets:

 Basic Petri Nets: These are the foundation for modeling concurrent systems.

 Colored Petri Nets (CPNs): These extend basic Petri nets by allowing tokens to carry data
values, making them suitable for modeling complex systems with data dependencies.

 Stochastic Petri Nets (SPNs): These incorporate timing information, allowing for modeling
and analysis of performance aspects in cloud systems.

5. Advantages of using Petri Nets:

 Formalism: Provides a mathematically rigorous framework for modeling and analyzing

concurrency.

 Graphical Representation: Makes it easier to understand and visualize complex system

interactions.

 Analysis Capabilities: Enables formal verification and performance analysis of cloud

systems.

In conclusion, Petri nets offer a