Complexity Language Engineering Domain-Specific Modelling (DSM) Conclusions. Multi-Paradigm Modelling and Language Engineering.

Transcription

1 Multi-Paradigm Modelling and Language Engineering Hans Vangheluwe Modelling, Simulation and Design Lab (MSDL) School of Computer Science, McGill University, Montréal, Canada CAMPaM Workshop 7 April 2009 Bellairs Research Institute, Barbados

2 Dealing with Complexity

3 Causes of complexity? Different Abstraction Levels (need morphism)

4 Causes of complexity? Large Number of Components

5 Causes of complexity? Multiple Formalisms: Power Window

6 Causes of complexity? Components in Different Formalisms

7 Causes of complexity? Multiple Views in Different Formalisms (need consistency!)

8 Causes of complexity? View: Structure Diagram

9 Causes of complexity? View: Events Diagram

10 Causes of complexity? View: Protocol Machine

11 Causes of complexity? Non-compositional/Emergent Behaviour

12 Causes of complexity? Multi-Paradigm Modelling ( minimize accidental complexity ) at the most appropriate level of abstraction using the most appropriate formalism(s) Differential Algebraic Equations, Petri Nets, Bond Graphs, Statecharts, CSP, Queueing Networks, Lustre/Esterel,... with transformations as first-class models Pieter J. Mosterman and Hans Vangheluwe. Computer Automated Multi-Paradigm Modeling: An Introduction. Simulation 80(9): , September Special Issue: Grand Challenges for Modeling and Simulation.

13 The Need for Transformations Waste Water Treatment Plants (WWTPs) NATO s Sarajevo WWTP

14 The Need for Transformations What does this WWTP model mean? settler influent aeration_tank mixer f_influent f_mixed f_processed f_out effluent f_bacteria

15 The Need for Transformations Transformation from WWTP to... influent f_influent f_influent mixer f_mixed f_mixed aeration_tank f_processed TRANSFORM C_influent 0.0 f_influent TRANSFORM f_influent f_bacteria f_mixed TRANSFORM f_mixed C_aeration 0.9 aeration_fraction f_processed OUT I OUT f_bacteria

16 The Need for Transformations... its meaning (steady-state abstraction): Causal Block Diagram (CBD) one 1.0 C_aeration C_settling 0.9 settling_fraction 0.6 OUT aeration_fraction f_influent f_mixed f_processed f_out negated dump_fraction C_influent 10.0 OUT I OUT I OUT effluent I OUT f_bacteria

17 C_influent 10.0 C_bacteria 1.0 f_influent f_bacteria aeration_fraction OUT f_mixed I C_aeration 0.9 OUT f_processed settling_fraction I OUT I C_settling 0.6 OUT f_dump f_out 1.0 effluent dump one negated OUT dump_fraction The Need for Transformations Meaning of the CBD = f _influent = C_influent f _bacteria = C_bacteria f _mixed = f _influent + f _bacteria aeration_fraction = C_aeration f _processed = aeration_fraction f _mixed settling_fraction = C_settling negated = settling_fraction one = 1 dump_fraction = one + negated f _dump = f _processed dump_fraction f _out = settling_fraction f _processed

18 The Need for Transformations Other uses of transformation operational semantics (as opposed to denotational) requirements design application requirements tests optimization... multi-formalism modelling

19 The Need for Transformations Multi-Formalism Modelling: Formalism Transformation Graph PDE KTG Bond Graph a-causal Cellular Automata System Dynamics DAE non-causal set Causal Block Diagram DAE causal set Bond Graph causal Transfer Function DAE causal sequence (sorted) scheduling-hybrid-dae Statecharts Activity Scanning Discrete Event Timed Automata DEVS&DESS Difference Equations DEVS state trajectory data (observation frame) Process Interaction Discrete Event Petri Nets 3 Phase Approach Discrete Event Event Scheduling Discrete Event

20 Modelling Language Engineering Dissecting a Modelling Language

21 Modelling Language Engineering A Production System Model

22 Modelling Language Engineering Modelling Languages as Sets of Models

23 Modelling Language Engineering: Meta-Modelling Meta-Modelling

24 Modelling Language Engineering: Meta-Modelling Modelling Abstract Syntax: Meta-Model not shown: local and global contraints

25 Modelling Language Engineering: Meta-Modelling Modelling Concrete Syntax (and UI Behaviour)

26 Modelling Language Engineering: Meta-Modelling Meta-Modelling Challenges scalability of (meta-)models meaningful model version control (meta-)model evolution deal with concrete syntax (mix textual/visual) in a unified manner

27 Modelling Language Engineering: Model Transformation Modelling Operational Semantics in the form of Rules Note the use of concrete syntax!

28 Modelling Language Engineering: Model Transformation Denotational Semantics

29 Modelling Language Engineering: Model Transformation Modelling Denotational Semantics

30 Modelling Language Engineering: Model Transformation How: Transformation Triple-Rules (bi-directional!)

31 Modelling Language Engineering: Model Transformation Model Transformation Challenges standardization/interoperability scalability (expressiveness and performance) automated testing (of models and model transformations) trace-ability (backward links), debugging from transformations to relationships (consistency) higher-order transformations

32 Sets and Transformations

33 Sets and Transformations

34 Sets and Transformations

35 Sets and Transformations Design-Space Exploration

36 WWTP Domain-Specific Modelling Environment Henk Vanhooren, Jurgen Meirlaen, Youri Amerlinck, Filip Claeys, Hans Vangheluwe, and Peter A. Vanrolleghem. WEST: Modelling biological wastewater treatment. Journal of Hydroinformatics, 5(1):27-50, 2003.

37 Why DS(V)M? (as opposed to General Purpose modelling) match the user s mental model of the problem domain maximally constrain the user (to the problem at hand) easier to learn avoid errors separate domain-expert s work from analysis/transformation expert s work Anecdotal evidence of 5 to 10 times speedup Steven Kelly and Juha-Pekka Tolvanen. Domain-Specific Modeling: Enabling Full Code Generation. Wiley 2008.

38 DS(V)M example application, the PhoneApps Domain-Specific model

39 DS(V)M example application: conference registration (Google Android)

40 Only transform... ConferenceRegistrationApps 1 PhoneApps 2 3 StateCharts AndroidAppScreens 4 5 AndroidAppFiles 6 Actual files (AndroidManifest.xml, PhoneApp.java, PhoneAppStateChart.java, screen_*.xml)

41 Eat Your Own Dogfood!

42

43 Model Everything! (to deal with Change) at most appropriate level of abstraction using most appropriate formalism(s) with transformations as first-class models