Klaus Krogmann

Using Genetic Search for Reverse Engineering of Parametric Behavior Models for Performance Prediction

In component-based software engineering, existing components are often reused in new applications. Correspondingly, the response time of an entire component-based application can be predicted from the execution durations of individual component services. These execution durations depend on the runtime behavior of a component which itself is influenced by three factors: the execution platform, the usage profile, and the component wiring. To cover all relevant combinations of these influencing factors, conventional prediction of response times requires repeated deployment and measurements of component services for all such combinations, incurring a substantial effort. This paper presents a novel comprehensive approach for reverse engineering and performance prediction of components. In it, genetic programming is utilized for reconstructing a behavior model from monitoring data, runtime bytecode counts, and static bytecode analysis. The resulting behavior model is parameterized over all three performance-influencing factors, which are specified separately. This results in significantly fewer measurements: The behavior model is reconstructed only once per component service, and one application-independent bytecode benchmark run is sufficient to characterize an execution platform. To predict the execution durations for a concrete platform, our approach combines the behavior model with platform-specific benchmarking results. We validate our approach by predicting the performance of a file sharing application.

CoCoME - The Common Component Modeling Example

The example of use which was chosen as the Common Component Modeling Example (CoCoME) and on which the several methods presented in this book should be applied was designed according to the example described by Larman in [1]. The description of this example and its use cases in the current chapter shall be considered under the assumption that this information was delivered by a business company as it could be in the reality. Therefore the specified requirements are potentially incomplete or imprecise.

Performance Prediction for Black-Box Components Using Reengineered Parametric Behaviour Models

In component-based software engineering, the response time of an entire application is often predicted from the execution durations of individual component services. However, these execution durations are specific for an execution platform (i.e. its resources such as CPU) and for a usage profile. Reusing an existing component on different execution platforms up to now required repeated measurements of the concerned components for each relevant combination of execution platform and usage profile, leading to high effort. This paper presents a novel integrated approach that overcomes these limitations by reconstructing behaviour models with platform-independent resource demands of bytecode components. The reconstructed models are parameterised over input parameter values. Using platform-specific results of bytecode benchmarking, our approach is able to translate the platform-independent resource demands into predictions for execution durations on a certain platform. We validate our approach by predicting the performance of a file sharing application.

Sustainability guidelines for long-living software systems

Economically sustainable software systems must be able to cost-effectively evolve in response to changes in their environment, their usage profile, and business demands. However, in many software development projects, sustainability is treated as an afterthought, as developers are driven by time-to-market pressure and are often not educated to apply sustainability-improving techniques. While software engineering research and practice has suggested a large amount of such techniques, a holistic overview is missing and the effectiveness of individual techniques is often not sufficiently validated. On this behalf we created a catalog of “software sustainability guidelines” to support project managers, software architects, and developers during system design, development, operation, and maintenance. This paper describes how we derived these guidelines and how we applied selected techniques from them in two industrial case studies. We report several lessons learned about sustainable software development.

Reverse Engineering Component Models for Quality Predictions

Legacy applications are still widely spread. If a need to change deployment or update its functionality arises, it becomes difficult to estimate the performance impact of such modifications due to absence of corresponding models. In this paper, we present an extendable integrated environment based on Eclipse developed in the scope of the Q-Impress project for reverse engineering of legacy applications (in C/C++/Java). The Q-Impress project aims at modeling quality attributes (performance, reliability, maintainability) at an architectural level and allows for choosing the most suitable variant for implementation of a desired modification. The main contributions of the project include i) a high integration of all steps of the entire process into a single tool, a beta version of which has been already successfully tested on a case study, ii) integration of multiple research approaches to performance modeling, and iii) an extendable underlying meta-model for different quality dimensions.

Reverse Engineering Software-Models of Component-Based Systems

An increasing number of software systems is developed using component technologies such as COM, CORBA, or EJB. Still, there is a lack of support to reverse engineer such systems. Existing approaches claim reverse engineering of components, but do not support composite components. Also, external dependencies such as required interfaces are not made explicit. Furthermore, relaxed component definitions are used, and obtained components are thus indistinguishable from modules or classes. We present an iterative reverse engineering approach that follows the widely used definition of components by Szyperski. It enables third-party reuse of components by explicitly stating their interfaces and supports composition of components. Additionally, components that are reverse engineered with the approach allow reasoning on properties of software architectures at the model level. For the approach, source code metrics are combined to recognize components. We discuss the selection of source code metrics and their interdependencies, which were explicitly taken into account. An implementation of the approach was successfully validated within four case studies. Additionally, a fifth case study shows the scalability of the approach for an industrial-size system.

An industrial case study on quality impact prediction for evolving service-oriented software

Systematic decision support for architectural design decisions is a major concern for software architects of evolving service-oriented systems. In practice, architects often analyse the expected performance and reliability of design alternatives based on prototypes or former experience. Model-driven prediction methods claim to uncover the tradeoffs between different alternatives quantitatively while being more cost-effective and less error-prone. However, they often suffer from weak tool support and focus on single quality attributes. Furthermore, there is limited evidence on their effectiveness based on documented industrial case studies. Thus, we have applied a novel, model-driven prediction method called Q-ImPrESS on a large-scale process control system consisting of several million lines of code from the automation domain to evaluate its evolution scenarios. This paper reports our experiences with the method and lessons learned. Benefits of Q-ImPrESS are the good architectural decision support and comprehensive tool framework, while one drawback is the time-consuming data collection.

Workload-aware System Monitoring Using Performance Predictions Applied to a Large-scale E-Mail System

Offering services in the internet requires a dependable operation of the underlying software systems with guaranteed quality of service. The workload of such systems typically significantly varies throughout a day and thus leads to changing resource utilisations. Existing system monitoring tools often use fixed threshold values to determine if a system is in an unexpected state. Especially in low load situations, deviations from the systems expected behaviour are detected too late if fixed value thresholds (leveled for peak loads) are used. In this paper, we present our approach of a workload-aware performance monitoring process based on performance prediction techniques. This approach allows early detections of performance problems before they become critical. We applied our approach to the e-mail system operated by Germanys largest e-mail provider, the 1&1 Internet AG. This case study demonstrates the applicability of our approach and shows its accuracy in the predicted resource utilisation with an error of mostly less than 10%.

The CACTOS Vision of Context-Aware Cloud Topology Optimization and Simulation

Recent advances in hardware development coupled with the rapid adoption and broad applicability of cloud computing have introduced widespread heterogeneity in data centers, significantly complicating the management of cloud applications and data center resources. This paper presents the CACTOS approach to cloud infrastructure automation and optimization, which addresses heterogeneity through a combination of in-depth analysis of application behavior with insights from commercial cloud providers. The aim of the approach is threefold: to model applications and data center resources, to simulate applications and resources for planning and operation, and to optimize application deployment and resource use in an autonomic manner. The approach is based on case studies from the areas of business analytics, enterprise applications, and scientific computing.

Modelling Layered Component Execution Environments for Performance Prediction

Software architects often use model-based techniques to analyse performance (e.g. response times), reliability and other extra-functional properties of software systems. These techniques operate on models of software architecture and execution environment, and are applied at design time for early evaluation of design alternatives, especially to avoid implementing systems with insufficient quality. Virtualisation (such as operating system hypervisors or virtual machines) and multiple layers in execution environments (e.g. RAID disk array controllers on top of hard disks) are becoming increasingly popular in reality and need to be reflected in the models of execution environments. However, current component meta-models do not support virtualisation and cannot model individual layers of execution environments. This means that the entire monolithic model must be recreated when different implementations of a layer must be compared to make a design decision, e.g. when comparing different Java Virtual Machines. In this paper, we present an extension of an established model-based performance prediction approach and associated tools which allow to model and predict state-of-the-art layered execution environments, such as disk arrays, virtual machines, and application servers. The evaluation of the presented approach shows its applicability and the resulting accuracy of the performance prediction while respecting the structure of the modelled resource environment.