Composition and construction of embedded software families [Elektronische Ressource] / von Danilo Beuche

Sujets

• Informatik

Informations

• Publié par: otto-von-guericke-universitat_magdeburg
• Publié le: 01 janvier 2004
• Nombre de lectures: 28
• Langue: English
• Poids de l'ouvrage: 1 Mo

Extrait

Composition and Construction of
Embedded Software Families
Dissertation
zur Erlangung des akademischen Grades
Doktoringenieur (Dr. Ing.)
angenommen durch die
Fakult t f r Informatik
der Otto-von-Guericke-Universit t Magdeburg
von Dipl.-Inform. Danilo Beuche
geb. am 2.12.1968 in Nauen
Gutachter:
Prof. Dr. Wolfgang Schr der-Preikschat
Prof. Dr. Reiner Dumke
Prof. Dr. J rg Nolte
Promotionskolloquium:
Magdeburg, den 19. Dezember 2003Acknowledgements
Finishing this dissertation would have been impossible without the help of many persons. I
would like to express my gratitude to all of them. However, it is impossible to name all of
my voluntary and sometimes involuntary helpers.
My special thanks go to Wolfgang Schr der-Preikschat who gave me opportunity and room
to develop my own ideas. His research group at Otto-von-Guericke-Universit t Magdeburg
was the perfect place for me and my research project. Beside my advisor I have to thank my
colleagues Guido Domnick, Andreas Gal, Jens Lauterbach, AndrØ Maa , Daniel Mahren-
holz, and Michael Schulze from this group. They frequently had to use the Consul program
which is a substantial result of this thesis. I am grateful that they kept sending bug reports
instead of deleting it from the disk right away when it still did not work as expected after
the nal bug x. Olaf and Ute Spinczyk, who were my rst colleagues in the group, had
not only to live with my chaos for about ve years, they also were rst time readers of my
drafts. Their comments and also their believe in this work were invaluable. Frank Behrens,
who always tried to motivate me, is hopefully pleased with the result of his efforts.
The students who helped me implementing my ideas, Bianca R diger, AndrØ Herms, and
Sascha R mke, did a wonderful job. I hope they will bene t from their experiences as much
as I did.
Many other persons from the group and from other places have to be named. I had the
opportunity to stay twice at the University of Santa Catarina, Brasil. My host there, Prof.
Antonio Augusto Fr hlich provided the appropriate environment to write large parts of the
dissertation there. J rgen Lehmann always provided me with disk space, IP addresses and
any other technical equipment I required. Martina Engelke, who was so unfortunate to share
the of ce with me for more than ve years, almost always ignored the high entropy of things
on on my desk. I never expressed it, but I really appreciated this!
The people at pure-systems with Holger Papajewski as their head deserve my thanks since
they bring my visions and ideas into real life. I am very fortunate to see this happen!
The two reviewers, Reiner Dumke and J rg Nolte, were so fast that I could nish the last
part of the dissertation in 2003. They allowed me to provide myself a very nice Christmas
present. Thanks for that!
3Last but not least I have to thank Birgit Schelm. She not only beared my sometimes strange
behavior throughout the complete the dissertation project but also read and proof-read the
manuscript at least twice and was always faster than me xing the problems. I would not
have nished this work without her.
4Contents
1 Introduction 13
1.1 Motivation and Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.2 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.3 Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2 Embedded Software 21
2.1 Embedded Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2 Related Software Technology Areas . . . . . . . . . . . . . . . . . . . . . 24
2.2.1 Other Application Domains . . . . . . . . . . . . . . . . . . . . . 25
2.2.1.1 High-Performance Computing . . . . . . . . . . . . . . 25
2.2.1.2 Distributed Systems . . . . . . . . . . . . . . . . . . . . 25
2.2.2 Software Development Technologies . . . . . . . . . . . . . . . . 27
2.2.2.1 Component-Based Systems . . . . . . . . . . . . . . . . 27
2.2.2.2 Object Orientation . . . . . . . . . . . . . . . . . . . . . 28
2.3 Software Reuse in Embedded Systems . . . . . . . . . . . . . . . . . . . . 31
2.3.1 The Two Dimensions of Reuse . . . . . . . . . . . . . . . . . . . . 31
2.3.2 Reuse Problems in Embedded Systems . . . . . . . . . . . . . . . 32
2.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3 Family-Based Software Development 37
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.2 Family-Based Software Development and Product-Line Engineering . . . . 37
3.3 F Software Dev Process . . . . . . . . . . . . . . . . 40
3.4 Domain Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.4.1 Domain Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.4.2 Domain Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.4.3 Implementation . . . . . . . . . . . . . . . . . . . . . . . 43
3.5 Application Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.5.1 Requirement Analysis . . . . . . . . . . . . . . . . . . . . . . . . 44
3.5.2 Application Design and Implementation . . . . . . . . . . . . . . . 44
3.6 Selected Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
5Contents
3.6.1 Domain-Speci c Languages-Based Methodologies . . . . . . . . . 45
3.6.1.1 Draco . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.6.1.2 FAST ? Family-Oriented Abstraction, Speci cation and
Translation . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.6.1.3 P3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.6.2 Feature Model-based Methodologies . . . . . . . . . . . . . . . . . 48
3.6.2.1 FODA ? Feature Oriented Domain Analysis . . . . . . 48
3.6.2.2 FORM ? Feature-Oriented Reuse Method . . . . . . . . 51
3.6.2.3 Other Approaches . . . . . . . . . . . . . . . . . . . . . 53
3.6.3 Decision Model-Based Methodologies . . . . . . . . . . . . . . . 53
3.6.3.1 Other Approaches . . . . . . . . . . . . . . . . . . . . . 54
3.6.4 Generic Methodologies . . . . . . . . . . . . . . . . . . . . . . . . 54
3.6.4.1 ODM ? Organizational Domain Model . . . . . . . . . 54
3.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4 Con guration Techniques in Domain Design and Implementation 59
4.1 Preprocessors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.2 Object-Oriented Languages . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.2.1 Inheritance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.2.2 Aggregation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.2.3 Uses-Relation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.2.4 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.3 Meta-programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.4 AOP ? Aspect-Oriented Programming . . . . . . . . . . . . . . . . . . . 66
4.5 Generators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.5.1 GenVoca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
5 Tool-Based Construction and Composition 73
5.1 Foundations of CONSUL . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.1.1 Software Family Hierarchy . . . . . . . . . . . . . . . . . . . . . . 76
5.1.2 Restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.2 The CONSUL Feature Modeling Language . . . . . . . . . . . . . . . . . 79
5.2.1 Feature Model Structure . . . . . . . . . . . . . . . . . . . . . . . 79
5.2.2 Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.2.3 Feature Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.2.4 Model Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.2.5 Feature Sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.2.6 Language Representation . . . . . . . . . . . . . . . . . . . . . . . 84
5.3 Feature-Based Problem Analysis . . . . . . . . . . . . . . . . . . . . . . . 85
5.3.1 Problem Domain De nition . . . . . . . . . . . . . . . . . . . . . 86
6Contents
5.3.1.1 Domain Attributes . . . . . . . . . . . . . . . . . . . . . 86
5.3.1.2 Embedded Speci cs . . . . . . . . . . . . . . . . . . . . 87
5.3.2 Feature Starter Set . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.3.3 From Problem to Model . . . . . . . . . . . . . . . . . . . . . . . 87
5.3.4 Building a Feature Model . . . . . . . . . . . . . . . . . . . . . . 88
5.3.5 Feature Types versus Feature Relations . . . . . . . . . . . . . . . 90
5.3.6 From Feature Relations to a Feature Model . . . . . . . . . . . . . 91
5.3.7 Single Domain Model Approach . . . . . . . . . . . . . . . . . . . 93
5.3.8 Multiple Model . . . . . . . . . . . . . . . . . . 94
5.3.8.1 Domain Decomposition . . . . . . . . . . . . . . . . . . 94
5.3.8.2 Domain Composition . . . . . . . . . . . . . . . . . . . 96
5.3.8.3 Hierarchical Domain Models . . . . . . . . . . . . . . . 96
5.3.9 Feature Deployment Model . . . . . . . . . . . . . . . . . . . . . 98
5.4 CONSUL Component Models . . . . . . . . . . . . . . . . . . . . . . . . 102
5.4.1 CONSUL Family Model Structure . . . . . . . . .