Top Banner
Introduction Approach Framework Matching Conclusions A Feature-based Framework for Model-Driven Engineering Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang Nordic Workshop on Model Driven Engineering NW-MoDE 2007 IT University of G¨ oteborg 27-29 August 2007 Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering
16
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

A Feature-based Framework forModel-Driven Engineering

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang

Nordic Workshop on Model Driven EngineeringNW-MoDE 2007

IT University of Goteborg27-29 August 2007

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 2: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Overview

1 Introduction

2 Approach

3 Framework

4 Matching

5 Conclusions

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 3: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Objective

Definition

Our objective is to assist an engineer in developing a system based on asoftware system family by offering domain knowledge for reuse.

We provide assistance in developing a system in:

Offering obligatory and optional features of a system family

Automated analysis of systems for identifying requirements

Providing generation rules for realising particular features

Generating realising models upon a feature configuration

Generating code based on these models

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 4: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Basics - Information Integration Systems (IIS)

Definition

An Information Integration System (IIS) provides a single read-accesspoint to a set of autonomous heterogeneous data sources.

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 5: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Basics - Software system families

Definition

Software systems that share a particular reference architecture and havecommon and variant features belong to a software system family.

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 6: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

MDA process of engineering an IIS application

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 7: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Domain Engineering vs. Application Engineering

Definition

Domain Engineering gathers domain knowledge reusable in any phasewhen engineering an application of a system family.

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 8: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Feature Modelling

Definition

Feature Models are used to describe commonness and variety of systemsbelonging to a system family.

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 9: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Patterns represent autonomous aspects of domainknowledge and design.

Pattern

Patterns structure both domain knowledge and application engineering

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 10: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Pattern-based application engineering

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 11: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Framework Architecture

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 12: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Proposing patterns by matching feature configurations

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 13: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Matching process

Matching feature configuration with specializations of the system family

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 14: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Matching of IIS configuration and IIS specialization

Spec. / Config. BOUND REMOVED UNDECIDEDBOUND 1 0 0.5

UNDECIDED 0.5 0.5 0.5REMOVED 0 1 0.5

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 15: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

Matching result

Quantified match result tells the best suited pattern specialization andinherited conflicts

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering

Page 16: A Feature-based Framework for Model-Driven Engineering

IntroductionApproach

FrameworkMatching

Conclusions

The framework supports:

MDA-driven Application Engineering + Domain Engineering +Continuous Software Engineering

Main properties:

Feature models specify properties of a system family

Patterns structure reusable domain knowledge

Support of all parts of the engineering (and evolution) process

requirements analysis: feature models & configurationsreverse engineering: system analysis / feature discoveryforward engineering: feature-driven generators (for applying patterns)

Susanne Busse, Helko Glathe, Thomas Stark, Jianhong Zhang A Feature-based Framework for Model-Driven Engineering