UNIT-3

PaaS

Evolution of Computing Paradigms and

Related Components

Computing paradigms have evolved significantly over

time, shaping the way computational resources are

utilized and managed. Below is a comprehensive

overview of different computing paradigms, including

their definitions, features, advantages, disadvantages,

real-world examples, and non-technical real-life

analogies.

1. Standalone Computing (Traditional

Computing)

Definition:

Standalone computing refers to a computing model where

a single computer operates independently, without relying

on external networks or shared resources.

Features:

• Single system usage

• No external resource sharing

• Locally stored data and applications

• Limited scalability

Advantages:

✅ Easy to manage and secure

✅ No dependency on external networks

✅ Cost-effective for personal use

Disadvantages:
❌ Limited computing power

❌ No resource sharing, leading to inefficiency

❌ Data is vulnerable to loss if the system fails

Examples:

• A personal desktop computer running applications independently

• A calculator used for mathematical computations

Non-Technical Analogy:

Imagine a person cooking all meals at home. They have to buy

ingredients, prepare food, and clean up without any external help.

2. Cluster Computing

Definition:

• Cluster computing is a paradigm where

multiple computers (nodes) work together

as a single system to improve processing

power and efficiency.

Features:

• Multiple computers linked together

• High availability and fault tolerance

• Used for high-performance applications

• Load balancing among nodes

Advantages:

✅ Increases computational power

✅ Redundant nodes provide fault tolerance

✅ Cost-effective compared to a single supercomputer

Disadvantages:

❌ Complex setup and management

❌ Requires efficient load balancing

❌ Network dependency for communication

Examples:

• Google Search infrastructure

• NASA’s supercomputing clusters

Non-Technical Analogy:

Imagine a group of people carrying a heavy object together. The weight is

distributed, making the task easier.

3. Grid

Computin

Definition:

Grid Computing is a

computing model in which

multiple geographically

distributed computers

work together to solve

complex problems by

sharing computational

power.

Key Features:

• Resource Sharing: Computers across different locations share

processing power.

• Heterogeneous Environment: Works across different hardware

and operating systems.

• Large-Scale Computation: Solves complex scientific,

mathematical, and engineering problems.

• Decentralized Control: No central control; tasks are distributed

dynamically.

Architecture

of Grid

Computing:

User Layer –

Users submit

tasks to the grid.

Resource

Layer –

Consists of

connected

nodes providing

computing

power.

Grid

Middleware

Layer –

Manages

resource

allocation and

scheduling.

Application

Layer –

Executes tasks

using the

allocated grid

resources.
Types of Grid Computing:

Computational Grid

– Focuses on

CPU-intensive

computations (e.g.,

weather modeling).

Data Grid –

Manages and

processes

large-scale datasets

(e.g., CERN’s Large

Hadron Collider).

Service Grid –

Provides distributed

services like

cloud-based AI

processing.

Advantages:

✅ High Computational Power – Uses multiple systems to solve

large-scale problems.

✅ Cost-Effective – Utilizes idle computers instead of investing in

supercomputers.

✅ Resource Optimization – Maximizes the use of existing computing

resources.

✅ Fault Tolerance – If one node fails, another takes over the task.

Disadvantages:

❌ High Network Dependency – Requires high-speed communication

between nodes.

❌ Security Issues – Data shared across different locations can be

vulnerable.

❌ Complex Setup and Management – Requires sophisticated software

for task distribution.

❌ Latency Issues – Communication between distant nodes can cause

delays.

•Real-World Examples:

• SETI@home – Uses volunteers' computers to analyze space signals for

extraterrestrial life.

• Folding@home – Uses distributed computing to study protein folding for

disease

research.

• CERN's Large Hadron Collider (LHC) – Analyzes massive amounts of particle

physics data.

•Non-Technical Analogy:

• A global team working on a massive jigsaw puzzle, where each person

works on

a small piece, and the final puzzle is assembled when all pieces are
completed.

4. Distributed Computing

Definition:

Distributed Computing is a computing paradigm in which multiple

computers or nodes, connected over a network, work together to

solve a problem by sharing resources, tasks, and computations.

Key Features:

• Resource Sharing: Computing power, memory, and data

storage are distributed across multiple machines.

• Concurrency: Multiple processes execute simultaneously.

• Fault Tolerance: If one node fails, others can take over the

task.

• Scalability: More nodes can be added to enhance

performance.

• Decentralization: No single point of failure as tasks are

distributed.

Layered Architecture

Layered Architecture in

distributed systems

organizes the system into

hierarchical layers, each

with specific functions and

responsibilities. This

design pattern helps

manage complexity and

promotes separation of

concerns

Layers and Their Functions

• Presentation Layer

• Function: Handles user interaction and presentation of data. It is

responsible for user interfaces and client-side interactions.

• Responsibilities: Rendering data, accepting user inputs, and sending

requests to the underlying layers.

• Application Layer

• Function: Contains the business logic and application-specific

functionalities.
• Responsibilities: Processes requests from the presentation layer, executes

business rules, and provides responses back to the presentation layer.

• Middleware Layer

• Function: Facilitates communication and data exchange between

different components or services.

• Responsibilities: Manages message passing, coordination, and

integration of various distributed components.

• Data Access Layer

• Function: Manages data storage and retrieval from databases or other

data sources.

• Responsibilities: Interacts with databases or file systems, performs

data queries, and ensures data integrity and consistency.

Peer-to-Peer (P2P) Model

• Peer-to-Peer (P2P) Architecture is a decentralized network design where

each node, or “peer,” acts as both a client and a server, contributing
resources and services to the network. This architecture contrasts with

traditional client-server models, where nodes

have distinct roles as clients or servers.

• In a P2P architecture, all nodes (peers) are equal participants in the

network, each capable of initiating and receiving requests.

• Peers collaborate to share resources, such as

files or computational power, without relying on a central server.

Key Features of Peer-to-Peer (P2P)

Architecture in Distributed Systems

• Decentralization

• Function: There is no central server or authority. Each peer operates

independently and communicates directly with other peers.

• Advantages: Reduces single points of failure and avoids central

bottlenecks, enhancing robustness and fault tolerance.

• Resource Sharing

• Function: Peers share resources such as processing power, storage

space, or data with other peers.

• Advantages: Increases resource availability and utilization across the

network.

• Scalability

• Function: The network can scale easily by adding more peers. Each new

peer contributes additional resources and capacity.

• Advantages: The system can handle growth in demand without requiring

significant changes to the underlying infrastructure.

• Self-Organization

• Function: Peers organize themselves and manage network connections

dynamically, adapting to changes such as peer arrivals and departures.

• Advantages: Facilitates network management and resilience without

central coordination

Client Server Architecture in Distributed Systems

• Client-Server Architecture is a foundational model in

distributed systems where the system is divided into two main components:
clients and servers. This architecture defines how tasks and services are

distributed across different entities within a network.

• In Client-Server Architecture, clients request services or resources, while

servers provide those services or

resources.

• The client initiates a request to the server, which processes the request
and returns the appropriate

response.
• This model centralizes the management of resources and services on the
server side, while the client side

focuses on presenting information and interacting with users

Data-Centic Architecture in Distributed

Systems

• Data-Centric Architecture is an architectural style that focuses on the

central management and utilization of data. In this approach, data is

treated as a critical asset, and the system is designed around data

management, storage, and retrieval processes rather than just the

application logic or user interfaces.

• The core idea of Data-Centric Architecture is to design systems where

data is the primary concern, and various components or services are

organized to support efficient data management and manipulation.

• Data is centrally managed and accessed by multiple applications or

services, ensuring consistency and coherence across the system

Service-Oriented Architecture (SOA) in

Distributed Systems

• Service-Oriented Architecture (SOA) is a design paradigm in distributed

systems where software components, known as “services,” are provided

and consumed across a network. Each service is a discrete unit that

performs a specific business function and communicates with other

services through standardized protocols.

• In SOA, the system is structured as a collection of services that are

loosely coupled and interact through well-defined interfaces. These

services are independent and can be developed, deployed, and

managed separately.

• They communicate over a network using standard protocols such as

HTTP, SOAP, or REST, allowing for interoperability between different

systems and technologies.

Event-Based Architecture in Distributed Systems

• It is an architectural pattern

where the flow of data and

control in a system is driven

by events. Components in

an EDA system

communicate by producing

and consuming events,

which represent state

changes or actions within

the system.

Types of Distributed Systems:

•Distributed Computing Systems – Multiple computers

working together (e.g., blockchain).

•Distributed Information Systems – Data distributed

across different locations (e.g., banking systems).

•Distributed Embedded Systems – Embedded devices

communicating in real-time (e.g., IoT).

Advantages:

✅ Improved Performance – Tasks are divided and executed in

parallel.

✅ Resource Efficiency – Utilizes available resources

effectively.

✅ Fault Tolerance – If one node fails, the system continues to

work.
✅ Scalability – Easily add more nodes to increase computing

power.

✅ Lower Cost – Uses multiple low-cost machines instead of a

single expensive system.

Disadvantages:

❌ Complexity in Management – Requires synchronization and

coordination among nodes.

❌ Security Risks – Data and communication between nodes

can be vulnerable to attacks.

❌ Network Dependency – High reliance on fast and stable

network connections.

❌ Data Consistency Issues – Ensuring all nodes have

up-to-date data can be challenging.

Real-World Examples:

• Google Search – Uses distributed computing to process billions of

queries.

• Bitcoin & Blockchain – A distributed network of nodes verifying

transactions.

• Netflix – Distributes streaming content across data centers

worldwide.

Non-Technical Analogy:

• A restaurant with multiple chefs in different kitchens preparing dishes

simultaneously and serving them as soon as they are ready. This

speeds up service rather than a single chef cooking everything alone.

5. Utility Computing

Definition:

Utility Computing is a computing model where resources like

processing power, storage, and applications are provided

on-demand, similar to utilities like electricity, water, or gas. A

pricing model within cloud computing where you pay only for the

computing resources used.

💡 Cloud Computing is like subscribing to Netflix – you get access to a

broad range of services (IaaS, PaaS, SaaS).

💡 Utility Computing is like paying for each movie separately – you

only pay for what you consume

Key Features:

• On-Demand Access: Users can request resources when

needed.

• Pay-Per-Use Model: Charges are based on actual usage.

• Scalability: Resources can scale up or down as required.

• Managed Services: The provider handles infrastructure,

updates, and maintenance.

Architecture of Utility Computing:

•Infrastructure as a Service (IaaS): Provides virtual

machines, storage, and networking (e.g., AWS EC2,

Google Compute Engine).

•Platform as a Service (PaaS): Provides a platform for

application development (e.g., Google App Engine).

•Software as a Service (SaaS): Provides software

applications on a subscription basis (e.g., Microsoft

Office 365).

Advantages:

✅ Cost Savings – Users pay only for what they use, reducing

upfront investment.
✅ Scalability – Resources can be increased or decreased

based on demand.

✅ Flexibility – Access services from anywhere with an internet

connection.

✅ Reduced Maintenance – Providers handle infrastructure

maintenance and updates.

Disadvantages:

❌ Security Concerns – Data is stored on third-party servers,

increasing the risk of breaches.

❌ Internet Dependency – Requires a stable internet

connection for access.

❌ Limited Control – Users have less control over

infrastructure and configuration.

❌ Vendor Lock-in – Switching providers can be difficult due to

dependency on specific services.

Real-World Examples:

• Amazon EC2 – Offers virtual machines for computing tasks.

• Google Cloud Platform – Provides storage, computing, and AI

services on-demand.

• Microsoft Azure – Hosts applications and services in the cloud.

Non-Technical Analogy:

• Using electricity at home: You don’t buy a power plant but pay

for the electricity you consume monthly.

6. Cloud Computing

• Definition:

• Cloud computing is a paradigm where computing resources

(storage, processing power, networking) are delivered over the

internet.

• Features:

• On-demand resource availability

• Pay-as-you-go pricing

• Scalable and flexible infrastructure

• Accessibility from anywhere

Advantages:

✅ Reduced infrastructure costs

✅ Scalable and flexible

✅ Easy accessibility and remote collaboration

Disadvantages:

❌ Security and privacy risks

❌ Internet dependency

❌ Downtime issues if the provider faces outages

Examples:

• Google Drive, Dropbox (Cloud storage)

• AWS Lambda (Serverless computing)

• Netflix (Streaming service hosted on AWS)

Non-Technical Analogy:

• Using a taxi instead of owning a car. You don’t have to worry about
maintenance or storage; you just pay when you need it.

7. Edge Computing

Definition:

• Edge computing is a decentralized computing paradigm

where data processing and computation happen closer to the

source of data (at the "edge" of the network) rather than relying
on a centralized cloud or data center.

Key Features:

✔ Processing at the Source – Computation happens near IoT

devices, sensors, or local servers.

✔ Low Latency – Reduces delays by processing data locally

instead of sending it to the cloud.

✔ Bandwidth Efficiency – Less data is sent over the internet,

reducing network congestion.

✔ Improved Security & Privacy – Sensitive data can be

processed locally without exposing it to external networks.

✔ Real-Time Processing – Ideal for applications that require

instant responses, like autonomous vehicles and smart cities.

Architecture of Edge Computing

Edge computing follows a layered architecture, typically including:

1. Edge Devices:

1. Sensors, IoT devices, smartphones, smart cameras.

2. Example: A self-driving car processing obstacle detection data in real-time.

2. Edge Nodes/Gateways:

1. Local servers or edge routers processing intermediate data.

2. Example: A factory’s edge server analyzing machine performance before

sending data to the cloud.

3. Edge Data Centers:

1. Small, distributed data centers placed closer to users/devices.

2. Example: Telecom companies deploying edge servers to improve 5G

connectivity.

4. Cloud/Data Center (Optional):

1. Used when deeper analysis or long-term storage is needed.

Advantages of Edge Computing

✅ Lower Latency: Faster processing as data doesn’t have to

travel to centralized cloud servers.

✅ Reduced Bandwidth Usage: Less data transmission means

lower costs and faster speeds.

✅ Improved Reliability: Works even with intermittent internet

connectivity.

✅ Enhanced Privacy & Security: Sensitive data stays local,

reducing cyberattack risks.

✅ Better Performance for IoT & AI Applications: Real-time

analytics for applications like healthcare, autonomous vehicles, and

industrial automation.

Disadvantages of Edge Computing

❌ Higher Initial Costs: Deploying local edge servers and

infrastructure can be expensive.

❌ Limited Processing Power: Edge devices have less

computing power than cloud data centers.

❌ Security Risks at the Edge: If not properly secured, edge

devices can be entry points for cyberattacks.

❌ Management Complexity: Requires maintenance of multiple

distributed devices instead of a single centralized cloud.

Real-World Examples of Edge Computing

1. Autonomous Vehicles (Self-Driving Cars 🚗)

• Edge computing processes sensor data (lidar, cameras, radar)

locally in milliseconds to make instant driving decisions.

• Without Edge Computing: If data had to be sent to the cloud first,

the delay could cause accidents.

2. Smart Cities (Traffic Management 🚦)

• Cameras and sensors at traffic signals process data locally to

adjust traffic lights dynamically based on congestion.

• Without Edge Computing: Sending all data to a central server

would create delays in decision-making.

3. Healthcare (Remote Patient Monitoring 🏥)

• Wearable health devices analyze patient vitals in real-time and send alerts
for abnormal readings.

• Without Edge Computing: Cloud processing delays could be critical in

emergencies.

4. Retail (Smart Checkout Systems 🛒)

• AI-powered checkout counters analyze shopping carts and complete

payments instantly.

• Without Edge Computing: Cloud-based systems may introduce lags,

frustrating customers.

5. Industrial Automation (Manufacturing 🏭)

• Edge servers in factories analyze machine sensor data to detect failures

before they occur.

• Without Edge Computing: Cloud-based analysis could be too slow to

prevent

damage.

8. Fog Computing

• Definition:

• Fog computing extends cloud computing by bringing

computation and storage closer to the edge of the network.

• Features:

• Decentralized data processing

• Reduces latency compared to traditional cloud computing

• Supports real-time analytics

• Advantages:

✅ Improved speed and performance

✅ Reduces reliance on cloud data centers

✅ Better bandwidth utilization

• Disadvantages:

❌ Security risks due to distributed nodes

❌ Complex architecture

❌ Higher maintenance cost

• Examples:

• Smart cities managing traffic sensors

• Healthcare monitoring systems processing patient data

• Non-Technical Analogy:

• A restaurant having small kitchens on each floor instead of a single large

kitchen for the whole building.

Platform as a service(PaaS)/ Cloud Platform

services

•Platform as a service (PaaS) is a cloud computing model

where a third-party provider delivers hardware and software

tools to users over the internet. Usually, these tools are needed

for application development.

• A PaaS provider hosts the hardware and software on its

own infrastructure. As a result, PaaS frees developers from

having to install in-house hardware and software to develop or

run a new application.

•PaaS tools tend to be touted as simple to use and convenient.

Users will normally have to pay on a per-use basis. An

organization may find the move to a PaaS compelling

considering potential cost savings over using on-premises

alternatives

How does PaaS work?

PaaS does not replace a company's entire IT infrastructure for

software development. It is provided through a cloud service

provider's hosted infrastructure.

Users most frequently access the offerings through a web

browser. PaaS can be delivered through public, private and

hybrid clouds to deliver services such as application hosting

and Java development on a per-use basis.

•Other PaaS services include: Development team

collaboration, Application design and development,

Application testing and deployment, Web service integration,

Information security, Database integration.

Advantages of PaaS

1. Faster Development & Deployment ⚡

• Developers can focus on coding rather than managing infrastructure.

• Pre-configured tools and frameworks speed up development.

• Reduces time-to-market for applications.

2. Cost-Effective 💰

• No need to invest in hardware, networking, and server maintenance.

• Pay-as-you-go pricing eliminates upfront costs.

• Reduces IT staff requirements for managing infrastructure.

3. Scalability & Flexibility 📈

• Easily scale applications up or down based on traffic demands.

• Supports automatic load balancing for optimal performance.

• Allows seamless integration with cloud storage and databases.

4. Simplified Development Environment 🛠

• Provides ready-to-use APIs, libraries, and development

frameworks.

• Compatible with multiple programming languages (Python,

Java, .NET, etc.).

• Reduces configuration complexity for developers.

5. Multi-User Collaboration 👥

• Teams can work on the same project simultaneously from

different locations.

• Cloud-based version control and collaboration tools simplify

teamwork.

• Ideal for agile development and DevOps practices.

6. Automatic Maintenance & Updates 🔄

• Cloud provider handles security patches, updates, and system

maintenance.

• Ensures applications run on the latest software versions.

• Reduces downtime and maintenance burden on businesses.

7. Security & Compliance 🔐

• Providers implement data encryption, authentication, and

security protocols.

• Compliance with regulations like GDPR, HIPAA, ISO 27001.

• Protects against data loss with regular backups.

Disadvantages of PaaS

1. Vendor Lock-in 🔒
• Migrating applications to another PaaS provider can be challenging.

• Each platform has specific tools, APIs, and configurations.

• Businesses may struggle with interoperability between different

platforms.

• 2. Limited Customization 🎛

• PaaS solutions come with pre-defined configurations.

• Restricted access to underlying infrastructure limits control.

• Some applications may require custom environments, which PaaS

may not support.

3. Security Risks ⚠

• Data is stored on third-party servers, raising security concerns.

• Unauthorized access or misconfigurations can lead to

vulnerabilities.

• Companies must ensure proper data encryption and access

controls.

4. Performance Issues 🚦

• Shared cloud resources may cause performance fluctuations.

• High traffic loads can impact application response times.

• Requires proper resource allocation to maintain stability.

5. Cost Uncertainty 💸

• While initially cost-effective, long-term usage costs may increase.

• Higher fees for premium features, integrations, and additional storage.

• Businesses need cost-monitoring tools to avoid overspending.

6. Compliance & Data Regulations 📜

• Data may be stored in different geographical locations, affecting

compliance.

• Organizations handling sensitive data must ensure regulatory

adherence.

• PaaS may not be ideal for highly regulated industries like

banking and healthcare.

Contd..

Integration of Private cloud and Public

cloud

The integration of a private cloud and a public cloud is known as

Hybrid Cloud Computing. It combines the security and control of

a private cloud with the scalability and cost-effectiveness of a

public cloud.

Key Aspects of Hybrid Cloud Integration

• Data and Application Portability – Ensures seamless movement

of data and applications between private and public

[Link] Connectivity – Uses VPNs, dedicated

connections, or SD-WAN for secure [Link]

Management – Provides centralized monitoring and control over

both [Link] Distribution – Runs critical

workloads on a private cloud while offloading less-sensitive

tasks to the public cloud.

Principal Technologies

• Key to cloud dynamic infrastructure is Virtualization layer-

Hypervisor allows multiple operating system instances to run

as guests on same server.

• Main drivers for cloud computing- cost, agility, time to market.

• Build infrastructure using cloud orchestrator, provisioning

engine.

• They sit on top of virtualization layer working on network,

server, storage.

Contd..

• It is a layer of software that

• (a) interacts with multiple servers

• (b)enables IT departments to pool resources together across

servers

• (c)defines standardized tiers of services called virtual compute

centers.

Uses: users share underlying cloud infrastructure in a secure

multi tenant environment, creates VM & users can login and self

provision respective workloads, transport API which allows cloud

administrator and users to interrelate with cloud infrastructure,

allow writing workflows to automate creation of cloud

infrastructure.

Contd..

• Virtualization foundation of cloud. Consists of physical hardware

with hypervisor on top of it.

• The cloud orchestrator and provisioning engine consist of one

or more cells that communicate with single database and offer a

Web portal.

• Using web portal cloud administrator create cloud infrastructure

resources and users’ self provision thus enabling IaaS.

Cloud Strategy

•This section provides guidance to define cloud strategy

and artefacts that capture architecture of a

cloud-enabled application.

•These artefacts are meant for implementation phase of

cloud for enabling an application.

•Implementation planning phase of cloud enables an

application lying between business strategy definition

for the adoption of cloud and design, development and

implementation phases of application.

Contd..

•It translates the business intent to a set of IT

requirements for cloud based application, deriving the

high level structures of cloud based application and

defining a roadmap for implementation of application.

•So primary input for the cloud implementation planning

phase comes from cloud strategy for the business.

Contd- Key steps in cloud

implementation planning

• Understand the cloud strategy

• Define the cloud application requirements

• Assess the cloud readiness

• Define high level cloud architecture

• Identify changes in management requirements.

• Develop a roadmap and implementation plan.

Contd…

The Roadmap for strategy and planning for cloud

deployment helps in:

•Identification of business justification for adopting the cloud.

•Starting the architectural work for cloud computing

•Application readiness and fitment based on workload.

•Workload assessement and recommendation for steps in

migration.

•Selection of correct solution and implementation.

•Acquiring required skills, bridging technology gap.

•Management of roles, responsibilities and

customer-relationship.

•Development of high level cloud roadmap and value

Cloud Design and Implementation using

SOA

•SOA – Service Oriented Architecture.

•It is an architectural design which includes collection of

services in a network which communicate with each

other. The complication of each service is not noticeable

to other service.

•The service is a kind of operation which is well defined,

self contained that provides separate functionality such

as checking customer account details, printing bank

statements etc which doesn’t depend on the state of

other services.

Contd..

Why to use SOA?

• Widely used in market which responds quickly and makes

effective changes according to market situations.

• SOA keeps the implementation details of subsystems a secret.

• Allows interaction of new channels with customers, partners and

suppliers.

• Authorizes the companies to select software or hardware of

their choice as it acts as platform independence.

Contd..

•Features: uses interfaces solves difficult integration

problems in large systems, communicates customers,

providers and suppliers with messages by using XML

schema, uses message monitoring to improve

performance and detects security attacks, resuses the

service- low management cost.

•Disadvantages: Requires high investment cost( large

investment on technology, development and human

resource), greater overhead when a service interacts

with another service(increase in response time), not

suitable for GUI –more complex as SOA requires heavy

data exchange.

Conceptual Model

• The highest level conceptual component model is the main

artefact that provides an abstract view of the design of a cloud

application to business stakeholders.

• Abstract view describes how business needs will be met by

cloud application components without delving into their technical

details.

• Components identified can be decomposed into futher layered

conceptual component structures to convey further details of the

respective components.

Contd..

• The cloud conceptual component model should contain the

following elements:

1. 2. The conceptual structure of the cloud application.

Dynamic interactions and dependencies between various

conceptual components.

3. Components that comprise the cloud services provided by the

application, each of which may be made up of

sub-components.

They are not converted to physical components but retained at

the conceptual level.

Cloud service defined

• Refer Page no 47.