LEAxDSML: Language Engineering for Analyzable Executable DSMLs

 

FWF Logo

LEAxDSML is a new research project conducted at TU Wien, Vienna, Austria by the Business Informatics Group under the lead of Tanja Mayerhofer. The project is funded by the Austrian Science Fund (FWF) under the grant number P 30525-N31. It is expected to start in January 2018 and will run for three years.

infoWe are hiring two researchers for this project!
Interested? Have a look at the job announcement.

Context

Model Driven Engineering (MDE) aims at reducing the accidental complexity associated with the development of complex software-intensive systems resulting from the large gap between the concepts used by domain experts to express their problems and the concepts provided by general-purpose programming languages to implement solutions. MDE addresses this problem through the use of domain-specific modeling languages (DSML) that enable domain experts to overcome the gap between problem space and solution space through suitable abstractions and transformations.

Executable domain-specific modeling languages (xDSML) are particularly useful for the specification of behavioral aspects of complex systems, such as their interactions with the external environment and realized business processes. Therewith, they support domain engineers in modeling complex system behaviors providing the basis for the performance of early analyses of emerging behaviors, a capability that is of uttermost importance in the development of complex software-intensive systems.

In order to be useful, xDSMLs have to be equipped with model analysis tools supporting domain engineers in comprehending, exploring, and analyzing modeled behaviors. In current practice, such tools are, however, hand-crafted in an ad hoc manner imposing significant efforts and costs on the development process.

Goals

The goal of the project LEAxDSML is to establish systematic language engineering methods for xDSMLs that support an automated generation of model execution infrastructures and model analysis tools.
In particular, the project pursues the following goals:

  1. Provide systematic language engineering methods that support xDSML engineers to identify, design and implement analyzability concerns of xDSMLs.
  2. Provide automation techniques for deriving from xDSML implementations customized model testing and model debugging tools.
  3. Provide fault localization mechanisms that allow domain engineers to trace back incorrect behaviors discovered in model testing or debugging to the fault-causing model elements.

LEAxDSML Framework

The developed xDSML language engineer methods, automation techniques for generating model testing and debugging tools, and fault localization mechanisms will be integrated into the xDSML language engineering framework LEAxDSML. This framework will be implemented on top of the language workbench GEMOC Studio.

The architecture of the LEAxDSML framework is depicted below. Based on the formal definition of an xDSML’s execution semantics and analyzability concerns developed by the language engineer, domain-specific stimuli, trace, and property languages are automatically derived allowing to interact with, represent, and reason about the execution of models conforming to the xDSML. These domain-specific languages provide the basis for the subsequent generation of execution-based model analysis tools customized for the considered xDSML. In particular, the framework will allow to automatically generate for an xDSML a domain-specific testing language and accompanying test engine, as well as a domain-specific debugging language and accompanying debugger. To facilitate the effective localization of faults discovered through the execution of tests or debugging sessions, the framework will also provide a dedicated fault localization component that traces back failing tests, or in general violated properties, to the failure cause providing dedicated views on the model and its execution that highlight those elements involved in the failure.

LEA-xDSML framework

Literature

The following are publications on previous research results relevant to the project:

Contact

Tanja Mayerhofer
(Principle Investigator of the project)
Business Informatics Group, TU Wien
mayerhofer@big.tuwien.ac.at

Acknowledgements

FWF Logo

The LEAxDSML project is funded by the Austrian Science Fund (FWF) under the grant number P 30525-N31.

© 2012 moliz by Vienna University of Technology Suffusion theme by Sayontan Sinha