Mobile Computing Overview

Q1. Mobile Architecture and Features

Mobile Computing enables users to access computing services wirelessly through portable devices
such as smartphones, tablets, laptops, and wearable devices while on the move.

Components of Mobile Computing

1. Mobile Hardware – Smartphones, tablets, laptops

2. Mobile Software – Mobile OS (Android, iOS), applications
3. Communication Network – Wi-Fi, Bluetooth, Cellular networks (3G, 4G, 5G)

Characteristics

1. Portability: Easy to carry devices

2. Connectivity: Wireless communication anytime, anywhere
3. Mobility: Continuous service while moving
4. Context Awareness: Uses location, time, and user activity

Advantages

1. Real-time access to information

2. Increased productivity and flexibility
3. Supports remote work and learning

Challenges

1. Security and privacy issues

2. Limited battery life
3. Network coverage and bandwidth limitations

Applications

1. Mobile banking and digital payments

2. GPS navigation and ride-sharing apps
3. Telemedicine and mobile health (mHealth)
4. E-learning and mobile commerce

Mobile Computing – Real-Time Examples

Example 1: Mobile Banking & UPI Payments

When you use Google Pay / PhonePe / Paytm:

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 1/11
• Mobile device connects to bank servers wirelessly
• Transaction is processed securely in real time
Real-world use: Cashless payments in India

Example 2: GPS Navigation (Google Maps)

• Uses mobile device + GPS satellites + internet
• Provides real-time traffic updates and directions

Real-world use: Ride-sharing apps like Ola, Uber

Example 3: Online Classes & Remote Work

• Students attend classes using mobile phones
• Employees work from anywhere using mobile apps
Real-world use: Zoom, Google Meet, MS Teams

Q2. Five Essential Characteristics of Cloud Computing

1. On-Demand Self-Service

Users can get computing resources like servers, storage, and network automatically whenever needed.
No need to contact the service provider.
The process is fully automated.

2. Broad Network Access

Cloud services are available over the internet.

They can be accessed using different devices like mobile phones, laptops, tablets, etc.
Standard methods are used to access the services.

3. Resource Pooling

The cloud provider’s resources are shared among multiple users.

Resources are assigned and reassigned based on user demand.
Users do not know the exact physical location of their resources.

Examples: storage, processing power, memory, bandwidth.

4. Rapid Elasticity

Resources can be increased or decreased quickly according to demand.

Scaling up and scaling down happens automatically.
To users, resources appear unlimited.

5. Measured Service
Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 2/11
Resource usage is monitored and measured automatically.
Users are charged based on their usage (pay-as-you-use model).
Both provider and consumer can track usage.

Q3. Cloud Computing Architecture

CLOUD COMPUTING ARCHITECTURE AND MANAGEMENT

1️⃣ Introduction to Cloud Computing

Cloud computing is a modern technology that provides services like storage, software, and servers over
the Internet.

→ Before understanding cloud deeply, we must first understand its architecture and management.

Architecture means the hierarchical structure (arrangement of layers and components) of cloud
computing.
→ It explains how different parts of cloud are organized and how they work together.

Anatomy means the internal structure (core working structure) of the cloud.
→ After understanding architecture, we study anatomy, network connection, management, and
application migration.

Cloud computing is completely Internet-dependent technology, so network connectivity is very

important.
Cloud Management deals with managing cloud resources and services properly.
→ It focuses on maintaining QoS (Quality of Service – performance, availability, reliability of
services).

Services in cloud are provided based on SLAs (Service Level Agreements – contract between user and
provider defining service quality and time).

Not all applications can directly move to cloud.

→ Application Migration (process of converting and shifting applications to cloud properly) is required.

CLOUD ARCHITECTURE (4 Layers)

Cloud architecture is divided into 4 connected layers, where each layer depends on the previous one.

🔹 Layer 1: User / Client Layer

This is the lowest layer where users start interaction with cloud.
→ Without users, cloud services have no purpose.

Users connect using devices like:

Thin client (device with very low processing power, depends on server)

Thick client (normal computer with enough processing power)

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 3/11
 Mobile or handheld devices

Cloud applications look like normal web applications.

→ But internally, they use cloud features like scalability and virtualization.
Therefore, this layer contains all client devices that access cloud services.

🔹 Layer 2: Network Layer

After the user layer, connection is needed.
→ This connection is provided by the Network Layer.
In Public Cloud, connection happens through the Internet.
→ Users do not know the exact physical location of cloud (abstract location).

In Private Cloud, connection may happen through LAN (Local Area Network).
Minimum bandwidth (internet speed) is required to access cloud services.
Important:
→ Network connection between user and cloud is NOT covered under SLA.

Thus, this layer enables communication between user and cloud.

🔹 Layer 3: Cloud Management Layer

Once connection is established, resources must be managed.
→ This is done by the Cloud Management Layer.
It includes:

Cloud Operating System (controls cloud resources)

Resource management software (handles scheduling and provisioning)

Optimization tools (improves server and storage usage)

Main functions:

Scheduling (assigning tasks properly)

Provisioning (providing resources when needed)

Server consolidation (efficient use of servers)
Governance (controlling and monitoring cloud operations)

This layer directly affects SLA.

→ Any delay or failure can cause SLA violation (breaking service agreement).

SLA violation results in penalty to service provider.

Examples:

Public Cloud: AWS (Amazon Web Services), Microsoft Azure

Private Cloud: OpenStack, Eucalyptus

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 4/11
So, this layer manages and controls the cloud services.

🔹 Layer 4: Hardware Resource Layer

This is the top and most important layer.
→ It contains actual physical resources.
It includes:

Data Centers (large buildings with many servers)

Storage systems

High-speed network devices

In public cloud:
→ Large data centers are used.

In private cloud:
→ Either data centers or high-configuration systems are used.
This layer strongly affects SLA.
→ If hardware fails or delays occur, service provider must pay penalty.
Data centers use:

High-speed networks

Efficient algorithms (step-by-step procedures to transfer data efficiently)

One cloud can use multiple data centers.

→ Multiple clouds can also share one data center.

Thus, this layer provides actual computing power.

🔹 Conclusion (Easy Summary to Remember)

Cloud architecture has 4 connected layers:

User Layer → Users access cloud

Network Layer → Connects users to cloud

Management Layer → Manages and controls services

Hardware Layer → Provides physical resources

All layers work together to provide QoS (Quality of Service) based on SLA (Service Level Agreement).

Q4. Distributed Computing and Characteristics

Distributed Computing is a computing paradigm in which multiple independent computers (nodes)
connected through a network work together to solve a problem or provide a service.

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 5/11
Each node has its own processor and memory, but they coordinate to function as a single system.
 Communication overhead

Distributed Computing
Distributed Computing is a computing paradigm in which multiple independent computers (nodes)
connected through a network work together to solve a problem or provide a service.
Each node has its own processor and memory, but they coordinate to function as a single system.

Tasks are divided into smaller subtasks

Subtasks are assigned to different machines

Machines communicate and coordinate over a network

Final result is obtained by combining outputs

Characteristics
Multiple autonomous computers

Resource sharing (CPU, memory, storage)

Concurrency (tasks execute simultaneously)

Scalability (easy to add more machines)

Fault tolerance (system continues even if one node fails)

Components of Distributed Computing

1. Nodes (Computers) – Perform computation
2. Network – Connects nodes (LAN/WAN/Internet)
3. Middleware – Manages communication and coordination
4. Distributed OS / Software – Controls execution

Types of Distributed Systems

1. Client–Server Systems
2. Peer-to-Peer Systems

3. Cluster Computing

4. Grid Computing
5. Cloud Computing

How Distributed Computing Works

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 6/11
 1. User submits a task

2. Task is split into subtasks

3. Subtasks are distributed to different nodes

4. Nodes process subtasks independently

5. Results are collected and combined

Real-Time Examples
1. Web Applications

Backend servers handle millions of users

Example: Google, Amazon, Facebook

2. Cloud Services

Storage and computing spread across data centers

Example: Google Drive, AWS, Azure

3. Online Banking Systems

Transactions processed across multiple servers

Ensures reliability and availability

4. Scientific Research

Distributed data analysis and simulations

Used in weather prediction and space research

Advantages
High reliability and availability
Better performance through parallelism

Cost-effective using commodity hardware

Easy scalability

Disadvantages
Network dependency

Security challenges
Complex system design

Data consistency issues

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 7/11
Q5. Explain Cloud Computing as Platform and Service

Cloud Computing as Platform and Service

🔹 First: The Three Service Models of Cloud Computing

Cloud computing mainly provides services in three models:

IaaS (Infrastructure as a Service) – Provides basic infrastructure like servers, storage, and
networking.

PaaS (Platform as a Service) – Provides a platform to develop, run, and manage applications.
SaaS (Software as a Service) – Provides ready-to-use software applications over the Internet.

Now, we will explain SaaS and PaaS in detail (5 points each) in simple English.

🔵 Cloud SaaS (Software as a Service)

SaaS means the user uses ready-made software provided by the cloud company.

1️⃣ What it provides

SaaS gives users access to applications that run on cloud infrastructure.
(Cloud infrastructure includes network, servers, operating systems, and storage.)

→ The user only uses the software, everything else is handled by the provider.

2️⃣ No control over infrastructure

The consumer (user) does not manage or control the underlying cloud system.

→ The provider manages servers, storage, updates, and security.

→ The user may only change limited settings (like profile settings).

3️⃣ Access through internet

Applications are accessed through:

Web browser (thin client – simple browser interface)

Program interface (API – Application Programming Interface)

Example: Web-based email like Gmail.

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 8/11
4️⃣ No installation required
The software is not installed on the user’s computer.

→ It runs online.
→ Updates and maintenance are done automatically by the provider.

5️⃣ Examples of SaaS applications

Typical SaaS applications include:

CRM (Customer Relationship Management – manages customer data)

Business Intelligence tools (analyze business data)

Online accounting software

So, SaaS is best for users who only want to use software without worrying about maintenance.

🟢 Cloud PaaS (Platform as a Service)

PaaS provides a platform where users can develop and deploy their own applications.

1️⃣ What it provides

PaaS allows users to create or upload their own applications using programming languages and tools
supported by the provider.
→ It gives a ready-to-use development environment.

2️⃣ Control level

The user does not manage the underlying infrastructure (servers, network, storage).

→ But the user controls their own applications and settings.

→ They can configure the application-hosting environment.

3️⃣ Ready-to-run framework

PaaS is a packaged development framework.
(Framework = a pre-built structure to develop software easily.)

→ Developers can focus only on coding.

4️⃣ Managed by provider

The PaaS provider manages:

Servers

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 9/11
 Networks

Storage
Scalability (ability to handle more users automatically)

Maintenance

→ The user pays only for the resources used.

5️⃣ Examples of PaaS

Examples include:

Google App Engine

Microsoft Azure Services

So, PaaS is best for developers who want to build applications without managing hardware.

🔹 Easy Difference to Remember

SaaS → Use ready-made software.

PaaS → Build and run your own software.

IaaS → Rent infrastructure (servers and storage).

Q6. Difference Between Parallel Computing and Distributed

Computing
Basis Parallel Computing Distributed Computing

A computing method where a single problem A computing method where multiple

is divided into smaller parts and processed independent computers (nodes) connected
Definition
simultaneously using multiple processors in by a network work together to solve a
the same system. problem.
Usually works within a single computer Works across multiple independent
System Type system (multicore CPU or multiprocessor computers connected through
system). LAN/WAN/Internet.

Can use shared memory (common memory Each node has its own separate memory
Memory
space). (distributed memory).

Processors communicate through shared Nodes communicate through a network

Communication
memory (faster communication). (slower than shared memory).

Tightly coupled system (processors closely Loosely coupled system (independent

Coupling
connected). computers connected via network).

Limited scalability (restricted by hardware Highly scalable (easy to add more

Scalability
capacity of one system). machines).

If one node fails, system can continue

Fault Tolerance If one processor fails, the system may stop.
(fault tolerance).

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 10/11
 Basis Parallel Computing Distributed Computing

Example GPU processing, matrix multiplication, AI Cloud services, online banking, web
Applications training, 3D rendering. applications, Google search engine.

Can be expensive (high-performance More cost-effective (uses multiple

Cost
hardware required). commodity machines).

Share resources and improve reliability

Main Goal Reduce execution time by parallel execution.
across multiple systems.

Printed using ChatGPT to PDF, powered by PDFCrowd HTML to PDF API. 11/11