Jean-Marc Jézéquel
Bio:
Jean-Marc Jézéquel is a Professor at the University of Rennes and a member of the DiverSE team at IRISA/Inria. Since 2021 he is Vice President of Informatics Europe. From 2012 to 2020 he was Director of IRISA, one of the largest public research lab in Informatics in France. In 2016 he received the Silver Medal from CNRS and in 2020 the IEEE/ACM MODELS career award. He was an invited professor at McGill University in 2022. Since Sept. 2023, he is a fellow of the Institut Universitaire de France (IUF).
He is the author of 4 books and of more than 300 publications in international journals and conferences. He was a member of the steering committees of the AOSD and MODELS conference series. He is currently Associate Editor in Chief of IEEE Computer and of the Journal on Software and System Modeling, as well as member of the editorial boards of the Journal on Software and Systems, and the Journal of Object Technology. He received an engineering degree from Telecom Bretagne in 1986, and a Ph.D. degree in Computer Science from the University of Rennes, France, in 1989.
Abstracts:
Taming Variability in Software Engineering: Past, Present & Future
Finding better ways to handle software complexity and variability is the holy grail for a significant part of the software engineering community, and especially for the Model Driven Engineering (MDE) one. To that purpose, plenty of techniques have been proposed, leading to a succession of trends in model based software development paradigms in the last decades. While these trends seem to pop out from nowhere, we claim that most of them actually stem from trying to get a better grasp on the variability of software. We revisit the history of Model Based Software Engineering trying to identify the main aspect of variability they wanted to address when they were introduced. We conclude on what are the variability challenges of our time, including variability of data leading to machine learning of models.
How Deep Variability Challenges Performance Modeling
Performance modeling used to be an easy thing in the early days of computing, when systems were rather simple. But modern appliances are increasingly complex pieces of hardware, firmware and software, each layer carrying evermore variability and thus bringing uncertainty. Even for a simple thing such as a video encoder, variability stems out from processors, firmware/OS, choice of encoding algorithm, its parameterization heuristics, the used compiler and its options, and of course user data (the video to be encoded). We call that "deep variability" and detail how it makes it difficult to carry on a priori performance modeling. We explore how alternative approaches based on machine learning could help with this problem.
Links:
