0.0 by ChatGPT

Chapter 6: Architectural Design

Architectural Design Activities

Definition

• Architectural Design Activities refer to the main tasks involved in designing a software system’s architecture.

Causes

• Not specified in notes

Goals / Objectives

• Break down the system into organized parts
• Define how components interact
• Establish control mechanisms

Importance

• Helps structure the system clearly
• Supports communication and control planning

Procedures

• 1. System Organization – Structuring the system into sub-systems
• 2. Modular Decomposition – Breaking sub-systems into modules
• 3. Control Modeling – Defining control relationships between modules

Advantages & Disadvantages

• Advantages:
  • Clear framework for design
  • Supports reuse and complexity management
• Disadvantages:
  • Not specified in notes

Impact / Effect

• Leads to better system planning and communication
• Enables reuse and easier maintenance

Examples

• Not specified in notes

Key Takeaways

• Architectural design includes system organization, modular decomposition, and control modeling
• It helps define structure, communication, and control
• It’s essential for building maintainable and scalable systems

---

Importance of Architectural Design

Definition

• Architectural Design is the process of defining a structured framework for a software system.

Causes

• Need for organized control and communication
• Requirement to identify sub-systems

Goals / Objectives

• Establish control and communication framework
• Identify sub-systems within the system

Importance

• Provides a high-level view of the system
• Helps stakeholders understand the system
• Ensures system meets requirements
• Enables reuse and negotiation
• Manages complexity

Procedures

• Not applicable to this topic

Advantages & Disadvantages

• Advantages:
  • High-level system presentation
  • Requirement validation
  • Reusable architecture
  • Design negotiation
  • Complexity management
• Disadvantages:
  • Not specified in notes

Impact / Effect

• Improves stakeholder communication
• Supports system analysis and reuse
• Facilitates design discussions
• Helps manage complex systems

Examples

• Not specified in notes

Key Takeaways

• Architectural design is crucial for communication, reuse, and complexity management
• It helps validate requirements and negotiate design decisions
• It provides a reusable and understandable structure

---

System Organization / Structuring

Definition

• System Organization is the process of structuring a system into sub-systems and defining their communication.

Causes

• Need to manage complexity
• Requirement to share data and distribute components

Goals / Objectives

• Identify sub-systems
• Define communication between sub-systems
• Enable distribution and data sharing

Importance

• Helps visualize system structure
• Supports independent development of sub-systems
• Enables reuse and scalability

Procedures

• Represent system in block diagrams
• Define sub-systems and their roles
• Establish communication and interfaces
• Choose an organization model:
  • Repository model
  • Client-server model
  • Layered model

Advantages & Disadvantages

• Advantages:
  • Clear separation of concerns
  • Easier maintenance and scalability
• Disadvantages:
  • Depends on model used (see below for model-specific pros/cons)

Impact / Effect

• Affects how data is shared and processed
• Influences system performance and flexibility

Examples

• Packing Robot System with sub-systems like:
  • Vision system
  • Object Identification System
  • Arm Controller
  • Gripper Controller
  • Packing Selection System
  • Conveyer Controller

Key Takeaways

• System organization defines sub-systems and their communication
• It uses models like repository, client-server, and layered
• Helps manage complexity and enables reuse

---

System Organization Models

Definition

• These are common styles used to organize software systems.

---

Repository Model

Definition

• A model where a central database is used to share data among sub-systems.

Causes

• Need to share large amounts of data

Goals / Objectives

• Centralize data management
• Allow sub-systems to access shared data

Importance

• Simplifies data access
• Reduces duplication

Procedures

• Use a central repository
• Connect sub-systems to the repository

Advantages & Disadvantages

• Advantages:
  • Centralized management
  • Sub-systems don’t need to know how data is produced
  • Efficient data sharing
• Disadvantages:
  • Hard to distribute
  • Sub-systems must agree on a data model
  • Data evolution is expensive
  • No specific management policies

Impact / Effect

• Enables integration but limits flexibility

Examples

• Integrated CASE Toolset with:
  • Program Editor
  • Report Generator
  • Code Generator
  • Design Analyzer
  • Project Repository

Key Takeaways

• Repository model uses a central database for sharing data
• It’s good for large data but hard to distribute
• Sub-systems must compromise on data models

---

Client-Server Model

Definition

• A model where clients request services from servers over a network.

Causes

• Need to distribute data and processing

Goals / Objectives

• Separate data and processing
• Allow multiple clients to access services

Importance

• Supports distributed systems
• Enables scalability and flexibility

Procedures

• Set up servers for specific services
• Connect clients to servers via a network

Advantages & Disadvantages

• Advantages:
  • Easy data distribution
  • Effective networking
  • Cheaper hardware
  • Easy server upgrades
• Disadvantages:
  • Different data models cause inefficient sharing
  • Redundant management
  • No central register of services

Impact / Effect

• Enables distributed access but may cause data inconsistency

Examples

• Film and Picture Library System with:
  • Catalogue Server
  • Video Server
  • Picture Server
  • Hypertext Server
  • Clients accessing via wide-bandwidth network

Key Takeaways

• Client-server model separates services and clients
• It’s scalable and flexible but may have data sharing issues
• Useful for distributed systems

---

Layered Model

Definition

• A model where the system is organized into layers, each providing services to the layer above.

Causes

• Need for structured development and clear interfaces

Goals / Objectives

• Support incremental development
• Limit impact of changes to adjacent layers

Importance

• Helps manage complexity
• Supports modular development

Procedures

• Organize system into layers:
  • Operating System Layer
  • Database System Layer
  • Object Management Layer
  • Configuration Management Layer

Advantages & Disadvantages

• Advantages:
  • Clear interfaces
  • Supports incremental development
  • Changes affect only adjacent layers
• Disadvantages:
  • Not specified in notes

Impact / Effect

• Makes system easier to maintain and extend

Examples

• Version Management System with 4 layers:
  • Configuration Management
  • Object Management
  • Database System
  • Operating System

Key Takeaways

• Layered model organizes systems into stacked layers
• It supports modularity and easier updates
• Each layer provides services to the one above

---

Modular Decomposition

Definition

• Modular Decomposition is the process of breaking down sub-systems into smaller, manageable modules.

Causes

• Need to simplify complex systems
• Requirement to organize components clearly

Goals / Objectives

• Decompose sub-systems into modules
• Group related components together
• Improve maintainability and clarity

Importance

• Makes systems easier to understand and manage
• Supports reuse and modular development

Procedures

• Identify sub-systems
• Break each sub-system into modules
• Apply decomposition strategies:
  • Object-Oriented Decomposition
  • Function-Oriented Decomposition

Advantages & Disadvantages

• Advantages:
  • Easier to manage and maintain
  • Supports reuse and clarity
• Disadvantages:
  • Not specified in notes

Impact / Effect

• Leads to better system structure
• Improves development and maintenance efficiency

Examples

• Not specified in notes (see sub-strategies below for examples)

Key Takeaways

• Modular decomposition breaks sub-systems into smaller modules
• It uses object-oriented or function-oriented strategies
• Helps manage complexity and improve clarity

---

Object-Oriented Decomposition

Definition

• A strategy that organizes modules based on objects and their interactions.

Causes

• Need to model real-world entities
• Desire for encapsulation and reuse

Goals / Objectives

• Group related data and operations into objects
• Model system behavior using object interactions

Importance

• Reflects real-world structure
• Supports encapsulation and reuse

Procedures

• Define objects with attributes and operations
• Connect objects based on relationships
• Example structure:
  • Client, Receipt, Invoice, Payment
  • Attributes: CustomerNo, Name, Address, etc.
  • Operations: Issue(), sendReminder(), acceptPayment(), sendReceipt()

Advantages & Disadvantages

• Advantages:
  • Natural mapping to real-world entities
  • Supports reuse and encapsulation
• Disadvantages:
  • Not specified in notes

Impact / Effect

• Improves system modularity and clarity
• Makes maintenance and extension easier

Examples

• Invoice Processing System with objects:
  • Client, Invoice, Payment, Receipt
  • Each object has attributes and operations

Key Takeaways

• Object-oriented decomposition uses objects to organize modules
• It reflects real-world entities and supports reuse
• Common in systems like invoice processing

---

Function-Oriented Decomposition

Definition

• A strategy that organizes modules based on system functions and processes.

Causes

• Need to focus on operations and tasks
• Desire to structure system by functionality

Goals / Objectives

• Break system into functional components
• Group related tasks into modules

Importance

• Clarifies system operations
• Supports task-based development

Procedures

• Identify main functions
• Create modules for each function
• Example structure:
  • Issue receipts, Read invoices, Identify payments, Issue reminders
  • Modules: Receipts, Invoices, Payments, Reminders

Advantages & Disadvantages

• Advantages:
  • Clear functional breakdown
  • Easy to understand system flow
• Disadvantages:
  • Not specified in notes

Impact / Effect

• Helps organize system by tasks
• Improves clarity and functional focus

Examples

• Invoice Processing System with modules:
  • Receipts, Invoices, Payments, Reminders
  • Functions: Issue receipts, Identify payments, etc.

Key Takeaways

• Function-oriented decomposition organizes modules by tasks
• It’s useful for systems focused on operations
• Common in invoice and payment systems

---

Control Modeling

Definition

• Control Modeling is the process of defining how sub-systems are controlled and how services are delivered at the right time and place.

Causes

• Need to manage control flow between sub-systems
• Requirement to deliver services correctly

Goals / Objectives

• Establish control relationships between sub-systems
• Ensure services are delivered efficiently
• Define control styles for different system types

Importance

• Helps coordinate sub-system operations
• Ensures timely and correct service delivery
• Supports system reliability and responsiveness

Procedures

• Choose control style:
  • Centralized Control
    • Call-Return Model
    • Manager Model
  • Event-Based Control
    • Broadcast Model
    • Interrupt-Driven Model

Advantages & Disadvantages

• Advantages:
  • Enables structured control flow
  • Supports both sequential and concurrent systems
• Disadvantages:
  • Depends on control style used (see below for model-specific pros/cons)

Impact / Effect

• Affects system responsiveness and coordination
• Determines how sub-systems interact and respond

Examples

• Not specified in notes (see sub-models below for examples)

Key Takeaways

• Control modeling defines how sub-systems are managed
• It uses centralized or event-based styles
• Ensures services are delivered correctly and on time

---

Centralized Control

Definition

• A control style where one sub-system has overall responsibility and starts/stops other sub-systems.

Causes

• Need for structured control in sequential or concurrent systems

Goals / Objectives

• Centralize control logic
• Coordinate sub-system operations

Importance

• Simplifies control flow
• Ensures predictable behavior

Procedures

• Use one main controller
• Apply either:
  • Call-Return Model for sequential systems
  • Manager Model for concurrent systems

Advantages & Disadvantages

• Advantages:
  • Clear control hierarchy
  • Suitable for predictable systems
• Disadvantages:
  • May limit flexibility
  • Not ideal for distributed systems

Impact / Effect

• Provides structured control
• May reduce responsiveness in dynamic environments

Examples

• Not specified in notes (see sub-models below)

Key Takeaways

• Centralized control uses one main controller
• It’s suitable for sequential or concurrent systems
• Ensures structured and predictable control flow

---

Call-Return Model

Definition

• A top-down subroutine model used in sequential systems.

Procedures

• Main program calls routines in sequence
• Example structure:
  • Main Program
    • Routine 1, Routine 2, Routine 3
    • Routine 1.1, Routine 1.2, etc.

Examples

• Sequential systems with structured subroutine calls

Key Takeaways

• Call-return model is used for sequential systems
• It follows a top-down structure
• Each routine is called in order by the main program

---

Manager Model

Definition

• A model where one system component acts as a manager and controls other processes.

Procedures

• Designate a system manager
• Allow sub-systems to execute concurrently
• Example components:
  • Sensor Processes, Computation Processes, System Controller, Actuator Processes, User Interface, Fault Handler

Examples

• Concurrent systems with a central controller

Key Takeaways

• Manager model is used for concurrent systems
• One component manages all others
• Processes can run at the same time

---

Event-Based Control

Definition

• A control style where sub-systems respond to events from other systems or the environment.

Causes

• Need for dynamic and responsive control
• Requirement to handle external or internal events

Goals / Objectives

• Enable sub-systems to react to events
• Support distributed and real-time systems

Importance

• Improves responsiveness
• Supports integration across networks

Procedures

• Use either:
  • Broadcast Model
  • Interrupt-Driven Model

Advantages & Disadvantages

• Advantages:
  • Flexible and dynamic
  • Suitable for distributed systems
• Disadvantages:
  • May be harder to manage and predict

Impact / Effect

• Enables real-time responses
• Supports integration across multiple systems

Examples

• Not specified in notes (see sub-models below)

Key Takeaways

• Event-based control responds to system or environmental events
• It’s flexible and suitable for real-time systems
• Uses broadcast or interrupt-driven models

---

Broadcast Model

Definition

• A model where events are broadcast to all sub-systems, and those able to handle them respond.

Procedures

• Broadcast event to all sub-systems
• Sub-systems respond if capable
• Example structure:
  • Event and Message Handler
  • Subsystems 1–4 respond to events E1–E4

Examples

• Distributed systems across networks

Key Takeaways

• Broadcast model sends events to all sub-systems
• Only relevant sub-systems respond
• Useful for integrating distributed systems

---

Interrupt-Driven Model

Definition

• A model where sub-systems respond to interrupts, often used in real-time systems.

Procedures

• Use an interrupt handler
• Sub-systems respond when interrupted

Examples

• Real-time systems requiring immediate response

Key Takeaways

• Interrupt-driven model uses interrupts to trigger responses
• It’s suitable for real-time systems
• Ensures fast and targeted control

---

Final Revision Summary

System Organization

• Structure system into sub-systems
• Identify communication paths
• Models: Repository, Client-Server, Layered

Modular Decomposition

• Break sub-systems into modules
• Strategies: Object-Oriented, Function-Oriented

Control Modeling

• Define control relationships
• Styles: Centralized (Call-Return, Manager), Event-Based (Broadcast, Interrupt-Driven)