Matthias Kowal

Explaining anomalies in feature models

The development of variable software, in general, and feature models, in particular, is an error-prone and time-consuming task. It gets increasingly more challenging with industrial-size models containing hundreds or thousands of features and constraints. Each change may lead to anomalies in the feature model such as making some features impossible to select. While the detection of anomalies is well-researched, giving explanations is still a challenge. Explanations must be as accurate and understandable as possible to support the developer in repairing the source of an error. We propose an efficient and generic algorithm for explaining different anomalies in feature models. Additionally, we achieve a benefit for the developer by computing short explanations expressed in a user-friendly manner and by emphasizing specific parts in explanations that are more likely to be the cause of an anomaly. We provide an open-source implementation in FeatureIDE and show its scalability for industrial-size feature models.

Implicit constraints in partial feature models

Developing and maintaining a feature model is a tedious process and gets increasingly difficult with regard to large product lines consisting of thousands of features and constraints. In addition, these large-scale feature models typically involve several stakeholders from different domains during development and maintenance. We aim at supporting such stakeholders by deriving and explaining implicit constraints for partial feature models. A partial feature model can either be a submodel of a feature model representing the full product line or a specific feature model in a set of interrelated models. For every implicit constraint, we generate an explanation exposing which other model parts and constraints interfere with the partial model of interest. Thus, stakeholders are only confronted with a small part of the feature model reducing the complexity while preserving the necessary information about dependencies. Our approach is implemented in the open-source framework FeatureIDE.

Family-Based Performance Analysis of Variant-Rich Software Systems

We study models of software systems with variants that stem from a specific choice of configuration parameters with a direct impact on performance properties. Using UML activity diagrams with quantitative annotations, we model such systems as a product line. The efficiency of a product-based evaluation is typically low because each product must be analyzed in isolation, making difficult the re-use of computations across variants. Here, we propose a family-based approach based on symbolic computation. A numerical assessment on large activity diagrams shows that this approach can be up to three orders of magnitude faster than product-based analysis in large models, thus enabling computationally efficient explorations of large parameter spaces.

Scaling Size and Parameter Spaces in Variability-Aware Software Performance Models (T)

In software performance engineering, what-if scenarios, architecture optimization, capacity planning, run-time adaptation, and uncertainty management of realistic models typically require the evaluation of many instances. Effective analysis is however hindered by two orthogonal sources of complexity. The first is the infamous problem of state space explosion -- the analysis of a single model becomes intractable with its size. The second is due to massive parameter spaces to be explored, but such that computations cannot be reused across model instances. In this paper, we efficiently analyze many queuing models with the distinctive feature of more accurately capturing variability and uncertainty of execution rates by incorporating general (i.e., non-exponential) distributions. Applying product-line engineering methods, we consider a family of models generated by a core that evolves into concrete instances by applying simple delta operations affecting both the topology and the models parameters. State explosion is tackled by turning to a scalable approximation based on ordinary differential equations. The entire model space is analyzed in a family-based fashion, i.e., at once using an efficient symbolic solution of a super-model that subsumes every concrete instance. Extensive numerical tests show that this is orders of magnitude faster than a naive instance-by-instance analysis.

Delta modeling for variant-rich and evolving manufacturing systems

Manufacturing systems exist in many different variants and evolve over time in order to meet changing requirements or environment contexts. This leads to an increased design complexity as well as to increased maintenance effort. In order to appropriately handle this inherent complexity, we propose a multi-perspective modeling approach combining UML activity, component-based and state chart diagrams to separately represent different system aspects. We combine the multi-perspective modeling approach with delta modeling to capture the variability and evolution of these manufacturing systems. Delta modeling allows a flexible, yet concise and understandable representation of variability in a modular manner. We examine our approach by applying it to a manufacturing lab demonstrator system with automated code generation from models obtained by delta application.

Towards efficient SPL testing by variant reduction

Testing software systems plays a pivotal role for quality, reliability, and safety of such systems. Several approaches exist that provide efficient algorithms to test one software system. However, in the context of variable software systems, called software product lines (SPLs), testing has to deal with potentially thousands of variants. Unfortunately, current approaches do not scale to this problem and thus testing SPLs efficiently is a challenging task. In this paper, we propose an approach to reduce the test set by explicitly modeling information about shared resources and communication in feature models. As a result, we can figure out features that interact with each other and thus are more likely to cause problems. We show with a small case study that our approach reduces both, the features under test as well as the time for computing all feature combinations to be tested.

Selected challenges of software evolution for automated production systems

Automated machines and plants are operated for some decades and undergo an everlasting evolution during this time. In this paper, we present three related open evolution challenges focusing on software evolution in the domain of automated production systems, i.e. evolution and co-evolution of (interdisciplinary) engineering models and code, quality assurance as well as variant and version management during evolution.

Higher-order delta modeling for software product line evolution

In software product lines (SPL), i.e., a family of similar software systems sharing common and variable artifacts, modeling evolution and reasoning about it is challenging, as not only a single system, but rather a set of system variants as well as their interdependencies change. An integrated modeling formalism for variability and evolution is required to allow the capturing of evolution operations that are applied to SPL artifacts, and to facilitate the impact analysis of evolution on the artifact level. Delta modeling is a flexible transformational variability modeling approach, where the variability and commonality between variants are explicitly documented and analyzable by means of transformations modeled as deltas. In this paper, we lift the notion of delta modeling to capture both, variability and evolution, by deltas. We evolve a delta model specifying a set of variants by applying higher-order deltas. A higher-order delta encapsulates evolution operations, i.e., additions, removals, or modifications of deltas, and transforms a delta model in its new version. In this way, we capture the complete evolution history of delta-oriented SPLs by higher-order delta models. By analyzing each higher-order delta application, we are further able to reason about the impact and, thus, the changes to the specified set of variants. We prototypically implement our formalism and show its applicability using a system from the automation engineering domain.

Model-Based Testing

Software more and more pervades our everyday lives. Hence, we have high requirements towards the trustworthiness of the software. Software testing greatly contributes to the quality assurance of modern software systems. However, as todays software system get more and more complex and exist in many different variants, we need rigorous and systematic approaches towards software testing. In this tutorial, we, first, present model-based testing as an approach for systematic test case generation, test execution and test result evaluation for single system testing. The central idea of model-based testing is to base all testing activities on an executable model-based test specification. Second, we consider model-based testing for variant-rich software systems and review two model-based software product line testing techniques. Sample-based testing generates a set of representative variants for testing, and variability-aware product line testing uses a family-based test model which contains the model-based specification of all considered product variants.

1st International Workshop on UML Consistency Rules (WUCOR 2015): Post workshop report

The Unified Modeling Language (UML), with its 14 different diagram types, is the de-facto standard modeling language for object-oriented software modeling and documentation. Since the various UML diagrams describe different views of one, and only one, software system under development, they strongly depend on each other in many ways. In other words, the UML diagrams describing a software system must be consistent. Inconsistencies among these diagrams may be a source of faults during software development and analysis. It is therefore paramount that these inconsistencies be detected, analyzed and -- hopefully -- fixed. The goal of this workshop was to gather input and feedbacks on UML consistency rules from the community. This workshop provided an opportunity for researchers who have been working in the area of UML consistency to interact with each other at a highly interactive venue, improve the body of knowledge on UML consistency rules and discuss ideas for further research in this area. This report summarizes details of the workshop and the results obtained that day.