Unit 3: Software Design

Software Design

Software Design is the process of converting requirements (SRS) into a blueprint that guides developers in coding the software.

In simple words: Design = Planning how the software will be built

Objectives of Software Design

Convert requirements into workable solution
Reduce complexity
Improve software quality
Make coding and testing easier
Support maintenance and future changes

Design Levels

High-Level Design (Architectural Design)
Low-Level Design (Detailed Design)

Architectural Design (High-Level Design)

Architectural Design defines the overall structure of the software system, including major components and their interactions.

Key Elements

Subsystems / modules
Data flow between modules
Control relationships
External interfaces

Common Architectural Styles

Style	Description
Layered	System divided into layers
Client-Server	Client requests, server responds
Pipe & Filter	Data processed in stages
MVC	Model-View-Controller pattern

Advantages

Clear system structure
Easy to understand
Supports scalability
Improves maintainability

Low-Level Design (LLD)

Low-Level Design focuses on internal working of modules and detailed logic.

Modularization

Modularization is the process of dividing a system into smaller, manageable modules.

Each module performs a specific function

Advantages of Modularization

Easy debugging
Parallel development
Reusability
Better maintenance

Characteristics of a Good Module

High cohesion
Low coupling
Simple interface

Design Structure Charts (DSC)

A Design Structure Chart is a hierarchical diagram showing modules and their relationships.

Elements of Structure Chart

Element	Description
Rectangle	Module
Arrow	Data/control flow
Hierarchy	Parent-child modules

Importance

Visual representation of design
Helps in coding and testing
Shows module dependency

Pseudo Codes

Pseudocode is an informal, English-like description of program logic.

Features

Easy to understand
Not language-specific
Helps in algorithm design

Example


START
Read A, B
SUM = A + B
Print SUM
END

Flow Charts

A Flowchart is a graphical representation of program logic using standard symbols.

Common Flowchart Symbols

Symbol	Meaning
Oval	Start / End
Rectangle	Process
Diamond	Decision
Arrow	Flow direction

Advantages

Easy visualization
Helpful in debugging
Improves understanding

Coupling

Coupling refers to the degree of dependency between modules.

Lower coupling is better

Types of Coupling (From Best to Worst)

Type	Description
Data Coupling	Data passed as parameters
Stamp Coupling	Structure passed
Control Coupling	Control flags passed
Common Coupling	Shared global data
Content Coupling	One module modifies another

Cohesion

Cohesion refers to how closely related the elements inside a module are.

Higher cohesion is better

Types of Cohesion (From Worst to Best)

Type	Description
Coincidental	Unrelated tasks
Logical	Similar logic
Temporal	Tasks related to time
Procedural	Sequence based
Communicational	Same data used
Sequential	Output of one is input to another
Functional	Single, well-defined task

Coupling vs Cohesion (Very Important)

Basis	Coupling	Cohesion
Meaning	Inter-module dependency	Intra-module relation
Ideal	Low	High
Impact	Affects flexibility	Affects clarity
Quality	Low coupling → better design	High cohesion → better design

Quick Revision Table

Topic	Key Point
Software Design	Blueprint of software
Architectural Design	High-level structure
Modularization	Divide into modules
Structure Chart	Module hierarchy
Pseudocode	Logic description
Flowchart	Graphical logic
Coupling	Module dependency
Cohesion	Module internal strength

Important Exam Questions

Explain software design and its objectives
What is architectural design?
Explain modularization
Difference between coupling and cohesion
Types of cohesion and coupling

Conclusion

Software design transforms requirements into a structured solution. Good design focuses on modularization, low coupling, and high cohesion, resulting in high-quality, maintainable software.

Design Strategies in Software Engineering

Design Strategies are approaches or methods used to plan and structure software design so that the system becomes easy to understand, develop, test, and maintain.

Function-Oriented Design (FOD)

Function-Oriented Design focuses on what functions the system should perform. The system is divided into functions and sub-functions.

Emphasis is on processes, not data.

Key Features

Based on functions
Uses top-down approach
Data flows between functions
Common tools: DFD and Structure Charts

Steps in Function-Oriented Design

Identify main function
Divide into sub-functions
Create DFD
Develop structure chart

Advantages

Easy to understand
Suitable for small systems
Simple design approach

Disadvantages

Difficult to modify
Poor data security
Not suitable for large systems

Example

Payroll System
→ Calculate salary
→ Generate payslip
→ Deduct tax

Object-Oriented Design (OOD)

Object-Oriented Design organizes the system around objects, which contain data (attributes) and functions (methods).

Emphasis is on data and behavior together.

Key Features

Based on objects and classes
Uses encapsulation, inheritance, polymorphism
More flexible and reusable
Uses UML diagrams

Advantages

High reusability
Easy maintenance
Better data security
Suitable for large systems

Disadvantages

Complex design
Requires skilled developers
Higher initial cost

Example

Student Object

Data: Name, Roll No
Methods: Register(), PayFees()

Top-Down Design

Top-Down Design starts from the main system and gradually breaks it down into smaller modules.

From general to detailed

Steps

Identify main problem
Divide into sub-modules
Continue until simple modules

Advantages

Clear system structure
Easy planning
Better control

Disadvantages

Lower-level issues found late
Requires good initial understanding

Bottom-Up Design

Bottom-Up Design starts from small modules and integrates them to form a complete system.

Steps

Design basic modules
Combine modules
Form higher-level system

Advantages

Reusable components
Easy testing of modules
Suitable for object-oriented systems

Disadvantages

Difficult system planning
Architecture may not be clear initially

Comparison Tables (Very Important for Exams)

Function-Oriented vs Object-Oriented Design

Basis	Function-Oriented	Object-Oriented
Focus	Functions	Objects
Data Handling	Separate from functions	Encapsulated
Reusability	Low	High
Maintenance	Difficult	Easy
Suitable for	Small systems	Large systems

Top-Down vs Bottom-Up Design

Basis	Top-Down	Bottom-Up
Approach	General to detail	Detail to general
Start Point	Main system	Small modules
Planning	Easy	Difficult
Testing	Late	Early
Reusability	Low	High

Quick Revision Points

Function-Oriented Design → process-based
Object-Oriented Design → data + behavior
Top-Down → divide system
Bottom-Up → build system
Good design → low coupling, high cohesion

Important Exam Questions

Explain Function-Oriented Design
Compare Function-Oriented and Object-Oriented Design
Explain Top-Down and Bottom-Up Design
Advantages of Object-Oriented Design

Conclusion

Design strategies play a crucial role in developing quality software. Function-oriented design is simple and suitable for small projects, while object-oriented design is flexible and ideal for complex systems. Top-down and bottom-up approaches help structure the design process effectively.

Software Measurement and Metrics

Software Measurement is the process of assigning numbers or values to software attributes such as size, complexity, cost, and quality.

In simple words: Software Measurement = Measuring software to estimate effort, cost, and quality

What are Software Metrics?

Software Metrics are quantitative measures used to evaluate, predict, and improve software development and maintenance.

Objectives of Software Metrics

Estimate cost and effort
Measure software size and complexity
Improve quality
Control project progress

Size-Oriented Measures

Size-oriented measures evaluate software based on its size, usually calculated using:

Lines of Code (LOC)
Operators and operands
Functional features

Examples of Size-Oriented Metrics

Productivity = LOC / Person-month
Quality = Errors / KLOC
Cost = ₹ / LOC

Halstead’s Software Science

Halstead’s Software Science is a size-oriented metric that measures software complexity based on operators and operands used in the program.

Basic Terms in Halstead Metrics

Symbol	Meaning
n₁	Number of distinct operators
n₂	Number of distinct operands
N₁	Total operators
N₂	Total operands

Important Halstead Formulas

1. Program Vocabulary

$n = n_{1} + n_{2}$

2. Program Length

$N = N_{1} + N_{2}$

3. Volume

$V = N \times \log_{2} (n)$

4. Difficulty

$D = \frac{n_{1}}{2} \times \frac{N_{2}}{n_{2}}$

5. Effort

$E = D \times V$

Advantages of Halstead Metrics

Language independent
Simple mathematical model
Useful for effort estimation

Limitations

Ignores program structure
Not suitable for very large systems

Function Point (FP) Based Measures

Function Point Analysis measures software size based on functionality provided to the user, not on code length.

FP focuses on “what the software does”

Main Components of Function Points

Component	Description
External Inputs (EI)	User inputs
External Outputs (EO)	Reports, results
External Inquiries (EQ)	Online queries
Internal Logical Files (ILF)	Internal data
External Interface Files (EIF)	External data

Steps in Function Point Calculation

Step 1: Count Function Types

Assign complexity weights (Low, Average, High).

Step 2: Calculate Unadjusted Function Points (UFP)

Step 3: Calculate Value Adjustment Factor (VAF)

$V A F = 0.65 + (0.01 \times Total GSC)$

Step 4: Final Function Points

$F P = U F P \times V A F$

Advantages of FP Metrics

Independent of programming language
Suitable for early estimation
Better than LOC for large systems

Disadvantages

Complex calculation
Requires expert judgment

Cyclomatic Complexity Measure

Cyclomatic Complexity measures the logical complexity of a program by counting independent paths through the code.

Proposed by McCabe

Why Cyclomatic Complexity is Important

Measures code complexity
Helps in test case design
Indicates maintenance difficulty

Control Flow Graph (CFG)

A Control Flow Graph represents flow of control in a program.

Components of CFG

Component	Meaning
Node	Statement or block
Edge	Flow of control
Region	Area bounded by edges

Formulas for Cyclomatic Complexity (Exam Favourite)

Method 1: Using Edges and Nodes

$V (G) = E - N + 2$

Where:
E = Number of edges
N = Number of nodes

Method 2: Using Predicate Nodes

$V (G) = P + 1$

Where: P = Number of decision (if/while) nodes

Interpretation of Cyclomatic Complexity

Value	Meaning
1–10	Simple, low risk
11–20	Moderate complexity
21–50	High complexity
>50	Very complex, risky

Comparison Table (Exam-Oriented)

Metric	Focus
Halstead	Operators & operands
Function Point	User functionality
Cyclomatic Complexity	Control flow complexity

Important Exam Questions

Explain Halstead’s Software Science with formulas
What is Function Point Analysis? Explain steps
Define Cyclomatic Complexity
Explain Control Flow Graph
Compare LOC and FP metrics

Conclusion

Software measurement and metrics help in estimating effort, controlling complexity, improving quality, and managing software projects effectively. Metrics like Halstead, Function Points, and Cyclomatic Complexity are widely used in software engineering.