Steffen Zschaler

Ontologies, Meta-models, and the Model-Driven Paradigm

Ontologies are no silver bullet. They can be employed in the software process as descriptive standardized domain models, domain-specific languages, and modelling (description) languages. However, they should not be mingled with specifications of software systems. In MDE, both forms of models are needed and complement each other. It is time to develop appropriate mega-models that clarify the role of ontologies in MDE. This chapter has presented one approach; however, this can be only an intermediate step, because we restricted ourselves to the standard IRDS metapyramid. Other, more sophisticated meta-pyramids exist and must be extended to be ontology-aware.

Modular DSLs for Flexible Analysis: An e-Motions Reimplementation of Palladio

We address some of the limitations for extending and validating MDE-based implementations of NFP analysis tools by presenting a modular, model-based partial reimplementation of one well-known analysis framework, namely the Palladio Architecture Simulator. We specify the key DSLs from Palladio in the e-Motions system, describing the basic simulation semantics as a set of graph transformation rules. Different properties to be analysed are then encoded as separate, parametrised DSLs, independent of the definition of Palladio. These can then be composed with the base Palladio DSL to generate specific simulation environments. Models created in the Palladio IDE can be fed directly into this simulation environment for analysis. We demonstrate two main benefits of our approach: 1) The semantics of the simulation and the non-functional properties to be analysed are made explicit in the respective DSL specifications, and 2) because of the compositional definition, we can add definitions of new non-functional properties and their analyses.

The COMQUAD component model: enabling dynamic selection of implementations by weaving non-functional aspects

The reliability of non-functional contracts is crucial for many software applications. This added to the increasing attention this issue lately received in software engineering. Another development in software engineering is toward component-based systems. The interaction of both, non-functional aspects and components, is a relatively new research area, which the COMQUAD project is focusing on.Our component model, presented in this paper, enables the specification and runtime support of non-functional aspects in component-based systems. At the same time, a clear separation of non-functional properties and functionally motivated issues is provided. We achieve this by extending the concepts of the existing component-based systems Enterprise JavaBeans (EJB) and CORBA Components (CCM). Non-functional aspects are described orthogonally to the application structure using descriptors, and are woven into the running application by the component container acting as a contract manager. The container implicitly instantiates component specifications and connects them according to the current requests. The selection of actual implementations depends on the particular clients non-functional requirements. This technique also enables adaptation based on the specific quantitative capabilities of the running system.In this paper we give a detailed description of the COMQUAD component model and the appropriate container support. We also provide a simple case study of a multimedia application for better understanding.

VML* – a family of languages for variability management in software product lines

Managing variability is a challenging issue in software-product-line engineering. A key part of variability management is the ability to express explicitly the relationship between variability models (expressing the variability in the problem space, for example using feature models) and other artefacts of the product line, for example, requirements models and architecture models. Once these relations have been made explicit, they can be used for a number of purposes, most importantly for product derivation, but also for the generation of trace links or for checking the consistency of a product-line architecture. This paper bootstraps techniques from product-line engineering to produce a family of languages for variability management for easing the creation of new members of the family of languages. We show that developing such language families is feasible and demonstrate the flexibility of our language family by applying it to the development of two variability-management languages.

Finding the pattern you need: the design pattern intent ontology

Since the seminal book by the Gang of Four, design patterns have proven an important tool in software development. Over time, more and more patterns have been discovered and developed. The sheer amount of patterns available makes it hard to find patterns useful for solving a specific design problem. Hence, tools supporting searching and finding design patterns appropriate to a certain problem are required. To develop such tooling, design patterns must be described formally such that they can be queryed by the problem to be solved. Current approaches to formalising design patterns focus on the solution structure of the pattern rather than on the problems solved. In this paper, we present a formalisation of the intent of the 23 patterns from the Gang-of-Four book. Based on this formalisation we have developed a Design Pattern Wizard that proposes applicable design patterns based on a description of a design problem.

Formal specification of non-functional properties of component-based software systems

Component-based software engineering (CBSE) is viewed as an opportunity to deal with the increasing complexity of modern-day software. Along with CBSE comes the notion of component markets, where more or less generic pieces of software are traded, to be combined into applications by third-party application developers. For such a component market to work successfully, all relevant properties of components must be precisely and formally described. This is especially true for non-functional properties, such as performance, memory foot print, or security. While the specification of functional properties is well understood, non-functional properties are only beginning to become a research focus. This paper discusses semantic concepts for the specification of non-functional properties, taking into account the specific needs of a component market. Based on these semantic concepts, we present a new specification language QML/CS that can be used to model non-functional product properties of components and component-based software systems.

On Language-Independent Model Modularisation

As model-driven software development covers additional parts of the development process, the complexity of software models increases as well. At the same time, however, many modelling languages do not provide adequate support for modularising models. For this reason, there has been an increasing interest in the topic of model modularisation, often under the heading of aspect-oriented modelling (AOM). The approaches range from techniques that closely mimic concepts from aspect-oriented programming (AOP), such as AspectJ, to very powerful composition techniques for specific types of models--for example, state machines. We believe that AOM is more than just copying the concepts of AOP at the modelling level and should rightly include a large number of other model-composition techniques. However, developing model composition techniques and tooling is costly. To minimise the effort required, this paper presents a generic technique for model composition. The technique is based on invasive software composition and our Reuseware tooling and can be used with arbitrary modelling languages. The basic technique itself is language independent, but it can be adapted to construct language- and purpose-specific composition techniques for specific modelling languages and situations. Hence, it can be used both as a tool for developing specific model-modularisation techniques and as an instrument of research for studying basic properties and concepts of model modularisation. The paper gives a detailed description of our approach and evaluates it using a number of examples.

Relating feature models to other models of a software product line: a comparative study of featuremapper and VML

Software product lines using feature models often require the relation between feature models in problem space and the models used to describe the details of the product line to be expressed explicitly. This is particularly important, where automatic product derivation is required. Different approaches for modelling this mapping have been proposed in the literature. However, a discussion of their relative benefits and drawbacks is currently missing. As a first step towards a better understanding of this field, this paper applies two of these approaches-- FeatureMapper as a representative of declarative approaches and VML* as a representative of operational approaches--to the case study. We show in detail how the case study can be expressed using these approaches and discuss strengths and weaknesses of the two approaches with regard to the case study.

Domain-specific metamodelling languages for software language engineering

Domain-specific languages are constructed to provide modelling capabilities tailored to a specific domain. Sometimes, languages are developed many times, typically to support application in a new context. In doing so, recurring patterns and commonalities as well as variations across the evolving set of languages can be identified. This paper introduces the concept of a domain-specific metamodelling language, which codifies such commonalities and provides concepts and logic for expressing the variations. The challenges and difficulties of using domain-specific metamodelling languages are identified. We illustrate the concept with examples from different domains.

Rigorous identification and encoding of trace-links in model-driven engineering

Model-driven engineering (MDE) involves the construction and manipulation of many models of different kinds in an engineering process. In principle, models can be used in the product engineering lifecycle in an end-to-end manner for representing requirements, designs and implementations, and assisting in deployment and maintenance. The manipulations applied to models may be manual, but they can also be automated—for example, using model transformations, code generation, and validation. To enhance automated analysis, consistency and coherence of models used in an MDE process, it is useful to identify, establish and maintain trace-links between models. However, the breadth and scope of trace-links that can be used in MDE is substantial, and managing trace-link information can be very complex. In this paper, we contribute to managing the complexity of traceability information in MDE in two ways: firstly, we demonstrate how to identify the different kinds of trace-links that may appear in an end-to-end MDE process; secondly, we describe a rigorous approach to defining semantically rich trace-links between models, where the models themselves may be constructed using diverse modelling languages. The definition of rich trace-links allows us to use tools to maintain and analyse traceability relationships.