MOMA3N 2018 Abstracts


Full Papers
Paper Nr: 1
Title:

Integration of Hawk for Model Metrics in the MEASURE Platform

Authors:

Orjuwan Al-Wadeai, Antonio Garcia-Domínguez, Alessandra Bagnato, Antonin Abherve and Konstantinos Barmpis

Abstract: The MEASURE project aims to integrate metrics across all phases of the software development lifecycle into a single decision support platform. For the earlier phases, metrics can be derived from models. Industrial use of model-driven engineering produces large model repositories, and high-performance querying is key to keep their metrics up to date. This paper presents an integration between the MEASURE metrics platform and the Hawk model indexing tool. Hawk was improved in several ways, such as adding support for the new Modelio metamodelling framework, or allowing Hawk servers to be provisioned through configuration files rather than through its web services. MEASURE and Hawk were then combined successfully to extract metrics from Modelio models of various domains, and Hawk was able to index and efficiently answer queries about the 2GB collection of models used by Softeam to develop Modelio.

Paper Nr: 34
Title:

Detecting and Describing Variability-Aware Design Patterns in Feature-Oriented Software Product Lines

Authors:

Sven Schuster, Christoph Seidl and Ina Schaefer

Abstract: Software Product Lines (SPLs) enable customization by reusing commonalities and variabilities within a family of similar software systems. Design patterns are best practices of established solutions in object-oriented source code for recurring design challenges. Although certain design patterns realize variability, they are only defined in the context of stand-alone systems and not for SPLs. Employing design patterns to realize variability allows using best practices in design for SPL development. However, the exact usage of design patterns within SPLs has not been explored, and a formal notation to capture their usage within different features does not exist. In this work, we provide a model-based analysis method to determine the variability-aware usage of design patterns in source code within the context of Feature-Oriented Programming (FOP). Moreover, we introduce Family Role Models (FRMs) as an extension to role modeling, which offer a language-independent, unified, formal notation for decomposed design patterns. We apply the analysis method in a case study on the variability-aware usage of design patterns in feature-oriented SPLs and derive FRMs from the results.

Paper Nr: 35
Title:

Towards Automated Analysis of Model-Driven Artifacts in Industry

Authors:

Ramon Schiffelers, Yaping Luo, Josh Mengerink and Mark van den Brand

Abstract: Developing complex (sub)systems is a multi-disciplinary activity resulting in several, complementary models, possibly on different abstraction levels. The relations between all these models are usually loosely defined in terms of informal documents. It is not uncommon that only till the moment of integration at implementation level, shortcomings or misunderstanding between the different disciplines is revealed. In order to keep models consistent and to reason about multiple models, the relations between models have to be formalized. Multi-Disciplinary System Engineering (MDSE) ecosystems provide a means for this. These ecosystems formalize the domain of interest using Domain Specific Languages (DSLs), and formalize the relations between models by means of automated model transformations. This enables consistency checking between domain and aspect models and facilitates multi-disciplinary analysis of the single (sub)system at hand. MDSE ecosystems provide the means to analyze a single (sub)system model. A set of models of different (sub)systems can be analyzed to derive best modeling practices and modeling patterns, and to measure whether a MDSE ecosystem fulfills its needs. The MDSE ecosystem itself can be instrumented to analyze how the MDSE ecosystem is used in practice. The evolution of models, DSLs and complete MDSE ecosystems is studied to identify and develop means that support evolution at minimal costs while maintaining high quality. In this paper, we present the anatomy of MDSE ecosystems with industrial examples, the ongoing work to enable the various types of analysis, each with their dedicated purpose. We conclude with a number of future research directions.

Paper Nr: 36
Title:

Pain-mitigation Techniques for Model-based Engineering using Domain-specific Languages

Authors:

Benny Akesson, Jozef Hooman, Roy Dekker, Willemien Ekkelkamp and Bas Stottelaar

Abstract: Changing an established way of working can be a real headache. This is particularly true if there are high stakes involved, e.g., when changing the development process for complex systems. New design methods, such as model-based engineering (MBE) using domain-specific languages (DSLs) promise significant gains, such as cost reductions and improvements in productivity and product quality. However, transitioning between design methods comes with a great deal of uncertainty, as any approach has associated pains. While the gains may be intuitively appreciated, it may be less clear what the new pains will be and whether or not they will cancel out the gains. For this reason, it may sometimes feel safer to stick with the devil you know than to meet the one you do not, preventing the full design potential of the company from being reached. This paper is an experience report from an investigation into how to mitigate the pains associated with a transition to a model-based design flow using DSLs. The main contributions of the paper are: 1) a list of 14 pains related to MBE as a technology that is representative of our industrial partners designing high-tech systems in different domains, 2) a selected subset of six pains is positioned with respect to the state-of-the-practice, 3) practical experiences and pain-mitigation techniques from applying a model-based design process using DSLs to an industrial case study, and 4) a list of three open issues that require further research.

Short Papers
Paper Nr: 2
Title:

Towards Distributed Model Analytics with Apache Spark

Authors:

Önder Babur, Loek Cleophas and Mark van den Brand

Abstract: The growing number of models and other related artefacts in model-driven engineering has recently led to the emergence of approaches and tools for analyzing and managing them on a large scale. The framework SAMOS applies techniques inspired by information retrieval and data mining to analyze large sets of models. As the data size and analysis complexity goes up, however, further scalability is needed. In this paper we extend SAMOS to operate on Apache Spark, a popular engine for distributed Big Data processing, by partitioning the data and parallelizing the comparison and analysis phase. We present preliminary studies using a cluster infrastructure and report the results for two datasets: one with 250 Ecore metamodels where we detail the performance gain with various settings, and a larger one of 7.3k metamodels with nearly one million model elements for further demonstrating scalability.