Schedule

Abstract

Empirical studies in software engineering provide a systematic way of evaluating theories, languages, concepts, tools or methodologies, considering the industrial context in which they are applied. The course will prepare students by examining how to plan, conduct and report on empirical studies in software engineering. The course will cover all of the principal methods applicable to software engineering (controlled experiment, case studies, surveys, systematic literature reviews, and ethnography) and will describe quantitative and qualitative methods of analysis, including hypothesis testing and grounded theory. To showcase the different methods, the course will critically review representative examples of published work. At the end of the course, the students will be able to approach real-world research problems in a scientifically sound way, and contribute to theory building in software engineering research.

Schedule

Abstract

There is an ever-increasing need for more application performance and energy efficiency. However, since 2005 computing platforms are providing substantially higher performance only by increasing the number of processors (not anymore by just increasing clock frequency). Current processors range from multi-core encompassing even 128 cores to GP-GPU with more than 10000 cores on the single-chip and to the aggregation of these resources connected via high-speed interconnects. Parallel programming permits us to express an algorithm to properly exploit what the technology is providing us. But what are the principles of parallel programming that best allow (or limit) maximum performance? This course covers several foundational topics central to parallel computing, starting from the basic methods to properly identify performance bottlenecks and provides the advanced practical means to program a parallel machine by using the most used standards such as OpenMPI, OpenMP, CUDA. We analyze these concepts through several hands-on examples and real-world data.

Schedule

Abstract

Performance is a crucial aspect of software systems since it affects the user perception of the quality of service. It is relatively easy to measure performance indices such as utilization and response time from a running system; however, this information is insufficient to infer the behavior under unseen scenarios. To be able to carry out studies such as capacity planning and what-if analyses (e.g., what will be the response time if the number of users has a ten-fold increase?) and, crucially, to control performance targets at runtime automatically, one needs predictive models. This course will present techniques grounded on the theory of stochastic processes to build efficient and accurate performance models of software systems. In addition, it will present approaches to infer such models from a running system using approaches based on machine learning and optimization. Finally, applications to the performance prediction and control of microservices will be illustrated.

Schedule

Abstract

Schedule

Abstract

Schedule

Abstract

Schedule

Abstract

This course gives an introduction to multi-objective optimization (MOO) and explains how to develop and tune efficient multi-objective genetic algorithms (MOGAs). The course starts giving a mathematical background on MOO. The key concepts of genetic algorithms are then introduced for real, integer, mixed-integer, and combinatorial optimization problems, including some techniques to handle constraints. The focus then moves on some approaches to find the best hyperparameter values, and to carry out a performance evaluation based on typical measures of MOO. The course ends presenting some multi-criteria decision making techniques, with particular reference to choosing the final Pareto-optimal solution to implement.

Schedule

Abstract