System Modeling Quiz

Week 8 System Modeling and MDE

• System modeling is the process of creating abstract models of a system.
• Each model provides a different perspective of the system.
• Models are often presented using graphical notations, primarily the Unified Modeling Language (UML).
• System models are crucial for communicating system functionality to customers and for analysis by analysts.

System Perspectives

• External Perspective: Models the context and environment of the system.
• Internal Perspective: Models interactions within and between system components and its environment.
• Structural Perspective: Models the system's organization and data structure.
• Behavioral Perspective: Models how the system reacts to events.

Uses of Graphical Models

• Support discussion by facilitating feedback on existing or proposed systems, even if models are incomplete or incorrect.
• Documenting systems, acting as an accurate representation but not necessarily complete.
• Generating system implementation details.

UML Diagram Types

• Activity Diagrams: Illustrate activities (processes) within a system.
• Use Case Diagrams: Show interactions between a system and its environment.
• Sequence Diagrams: Depict interactions between actors and system components or actors to each other.
• Class Diagrams: Represent object classes and relationships within the system.
• State Diagrams: Illustrate how a system reacts to internal and external events.

Model-driven Engineering (MDE)

• Develops software from models rather than explicit programming code.
• Software automatically generated from models running on hardware or software platforms.
• Aims to raise the level of abstraction, reducing reliance on programming language details and execution program specifics.

Pros and Cons of MDE

• Pros: Allows for higher-level system abstraction, adaptable implementations across platforms, and potentially cheaper system adaptation.
• Cons: Models are not always optimal for implementation, and savings from generated code may be offset by the cost of developing translators.

Model Driven Architecture (MDA)

• A precursor to more general model-driven engineering (MDE).
• Uses a subset of UML models to describe a system.
• Models at varying levels of abstraction are created, with platform-independent models at higher levels.
• In principle, a working program can be generated without manual intervention.

4 Types of Models in MDA

• Computation-Independent Model (CIM): Focuses on domain abstractions, sometimes termed domain models.
• Platform-Independent Model (PIM): Models system operations for use across all potential implementations.
• Multiple Platform-Specific Models (PSMs): Transformations of the PIM for use on specific platforms (e.g. Java or .NET).

Adoption of MDA

• MDA is not easily consistent with agile methods, demanding modeling and planning.
• Specialized tool support for conversions and adaptation is needed for wide adoption.
• Limited tool availability and modifications required for organizational use are challenges.

Context and Interaction Models

• Context models illustrate the operational environment of a system, encompassing elements outside the system's boundaries.
• System boundaries are positioned by social and organizational considerations.
• Context models distinguish a system's context from how the system is used.
• Architectural models show relationships within the system and with external systems.

Process Perspective

• Process models detail how the system being developed is used within a wider business process.
• UML activity diagrams illustrate business process models.

Interation Models

• These models are important for identifying user needs, recognizing communication hurdles, and understanding system interaction structure.
• Use cases and sequence diagrams are common tools use in system modelling.

Structural Models

• Architectural structures (arrangements) of systems show components and relationships.
• Static models show the structure of the system at design.
• Dynamic models depict the system structure during system operation.

Class Diagrams

• A useful tool for object-oriented system design, revealing classes and their relationships.
• A class represents a general definition of an object, showing how different components in a system connect through these classes.

Behavior Models

• These model dynamic behavior of the system during operation.
• Models show what happens when a system reacts to changes in its environment, including responses to data and events.

Data-Driven Modeling

• Business systems that are primarily driven by data are often modelled using data-driven models.
• These models highlight ordered sequences of actions for processing input data and creating output in these systems

Event-driven Modeling

• Real-time systems frequently employ event-driven modeling where minimal data processing exists.
• Systems respond to internal and external stimuli (events), often represented using state machine models.
• State machine modeling illustrates the system's behavior with a finite number of states, transitioning between them based on external and internal events or stimuli.

Architectural Design Overview

• Architectural design determines a system's structure and overall organization.
• This is the bridge between the process of requirements engineering and the design process.
• Architectural representations use tools such as informal block diagrams to model system components and their relationships.

Architectural Views and Design Decisions

• Architectural design involves decisions about system structure, component interactions, and distribution.
• Multiple different views (e.g. logical, physical, development and process) are commonly employed.
• The architecture should align with functional requirements to determine the best structure.

System Characteristics in Architecture

• Factors such as performance, security, safety, availability and maintainability influence architectural decisions.
• Decisions regarding architecture should take these non-functional considerations into account.

Architectural Views

• Logical view identifies high-level abstractions in the system as objects or classes.
• Process view shows the interacting processes in the system and their communication during operation.
• Development view illustrates how the system is structured for software development.
• Physical view shows the system's hardware and how components are distributed across processors.
• Use cases or scenarios also play an important role in supplementing these 4 views.

Architectural Patterns

• Architectural Patterns are pre-defined, tried and tested structural models.
• Model–View–Controller (MVC) is a pattern for separating presentation, interaction, and data.
• Layer Architecture is a pattern that organizes system functionality into layers, with dependencies structured vertically.
• Repository Architecture is a pattern for managing data where information is maintained in a central repository.
• Client-Server Architecture distributes data and processing among interacting components (clients and servers).
• Pipe and Filter Architecture separates processing steps into individual, independent components for data analysis.

Additional Notes: (General points)

• UML is a common notation used
• ADLs are also used in system modeling
• The choice between models depends on the specifics of the system being modeled
• Multiple views are typically needed to fully understand a software system

This quiz evaluates your understanding of system modeling, including concepts of UML diagrams, model-driven engineering, and architectural perspectives. Test your knowledge on the purposes of graphical models and the implications of system structure in design discussions.