Autumn school

27th Sept – morning 9h00-10h30 coffee break  11h-12h30 Main Room CUI (ground floor) Speaker: Manuel Oriol, University of York, UK

Title: Testing Resilient Systems

Abstract: Under changing conditions, resilient systems are supposed to adapt and continue performing their task in a dutifully manner. Due to their dynamic nature, resilient systems are however difficult to test and validate. In this course we will review how traditional testing techniques can be applied to resilient systems and what are their limitations. We will also present challenges and research opportunities in this field.

Bio: Dr. Manuel Oriol is a senior lecturer at University of York on Large-Scale and Complex IT Systems. He has been lecturing on verification and validation techniques for the past 6 years. In particular, Manuel gave the Software Measurement and Software Testing module as well as the Safety-Critical Software Testing module at York for the past 3 years. Manuel also started the random testing project YETI.
Prior to being at University of York, Manuel worked at ETH Zurich on software engineering and AutoTest, University of Maryland on dynamic updates for C programs and graduated with a PhD from University of Geneva on dynamic updates for Java. Manuel also holds an Engineer Diploma from the ENSEEIHT school and an MSc from INPT.

27th Sept – afternoon 14h00-15h30 break 16h00-17h30 Main Room CUI (ground floor) Speaker: Valérie Issarny, INRIA Paris – Rocquencourt, France

Title: Towards Future Proof Interoperability

Abstract: The ubiquitous computing vision is hampered by the often extreme level of heterogeneity in the underlying infrastructure, which in turn impacts the ability to seamlessly interoperate. Further, the fast pace at which technology evolves at all abstraction layers increasingly challenges the lifetime of networked systems in the digital environment. As a matter of fact, existing middleware approaches and paradigms are unable to deliver on their most central promises, which is offering interoperability, in the increasingly heterogeneous networking environment. This talk discusses a promising approach to interoperability in ubiquitous environments, that is, synthesizing on the fly the protocols via which networked systems communicate. The supporting architectural framework together with the associated model-based approaches to the automated synthesis of protocols to sustain interoperability, from the application down to the middleware layers, will be presented. Further, the talk will demonstrate a prototype system that is developed as part of the European FET CONNECT project (http://connect-forever.eu/).

Bio: Dr. Valérie Issarny is “Directrice de recherche” at INRIA. Since 2002, she is the head of the ARLES INRIA research project-team at INRIA-Rocquencourt. Her research interests relate to distributed systems, software engineering, pervasive computing/ambient intelligence systems and middleware. She has (co)authored over 100 technical papers in the area of distributed systems and software engineering, and has been involved in a number of European and industrial projects. She has been and is serving as PC member, including as PC chair, of leading international events in the area of distributed systems, middleware, software engineering and trust management. She is currently the coordinator of the FP7 FET CONNECT project that revisits the middleware paradigm to sustain interoperability in the ever-changing pervasive networking environment. To know more about Valérie’s research, please visit https://www-roc.inria.fr/arles/members/issarny.html.

28th Sept – morning 9h00-10h30 coffee break  11h-12h30 Room 432 (3rd floor)

Speaker: Marc-Olivier Killijian, LAAS-CNRS, France

Title: Resilient Mobiquitous Systems: design, evaluation and geoprivacy issues

Abstract: Finding the proper abstractions to design middleware for the provision of dependable distributed applications on mobile devices remains a big challenge. The number of mobile communicating devices one can meet in every-day life is dramatically increasing: mobile phones, PDAs, handheld GPS, laptops and note- books, portable music and video players. Those devices benefit from an amazing number of sensors and communication interfaces. The interconnection of these systems does not only result in a huge distributed system. New technical and scientific challenges emerge due to the mobility of users and of their devices, or due to the massive scale of uncontrolled devices that constantly connect and disconnect, fail, etc. To handle those systems’ dynamics, cooperation-based approaches `a la peer-to-peer seem attractive. An important question we will try to answer during this talk is thus to know if and how can we design a sound middleware that offers useful building blocks for this type of system. Another crucial question we will discuss is how we can correctly evaluate those highly mobile and dynamic systems. Finally, we will address the privacy issues raised by the massive adoption of mobiquitous systems, in particular regarding the notion of geo-privacy.

Bio: Dr. Marc-Olivier Killijian received an Engineer degree from the National Institute of Applied Sciences of Toulouse in 1996, and a PhD in computer science from the National Polytechnic Institute of Toulouse in January 2000. His main interest at LAAS during this period concerned reflective computing, metaobject protocols and fault tolerance in distributed systems. Then he was working in the Trinity College of Dublin as a Research Fellow in the Distributed Systems Group, where he was involved in the definition and implementation of a new paradigm for location-aware group communication in ad-hoc mobile networks. He joined in 2001 the Dependable Computing and Fault-Tolerance Group at LAAS-CNRS as Chargé de Recherche. His research interests concern resilience issues in mobile and ubiquitous system : the impact of mobility at large, fault-tolerance and geo-privacy applied to various mobile systems (robots, automobiles, and other mobiquitous systems).

28th Sept – afternoon 14h00-15h30 break 16h00-17h30 Room 432 (3rd floor) Speaker: Ekkart Kindler, Technical University of Denmark

Title: Resilient Software by Model-based Software Engineering

Abstract: The idea of Model-based Software Engineering (MBSE) is to let models play a more important role in the software development, and make models more than just nice drawings — for example, by generating major parts of the code from models fully automatically. For some aspects this idea is working already today, and it is used in practice already. This way, software development can be much faster, less error prone, and more focussed on the conceptual problems. And since major parts of the software are generated fully automatically, model-based software engineering could be considered to be a means of software resilience.
The major part of this course, will be on the ideas and the basic concepts of Model-based Software Engineering, the state of the art and the limitation of today’s technologies — and on some visions on how to overcome these limitations. Though not put that way very often, MBSE can be considered as a means of software resilience, we will discuss MBSE’s potential and promises in this respect.

Bio: Ekkart Kindler is currently an associate professor in Computer Science and Engineering at the Technical University of Denmark (DTU). He received his masters and his PhD in Computer Science from the Technische Universtitaet Muenchen in 1990 and 1995, resp. He received his Habilitation in Computer Science from the Humboldt-Universitaet zu Berlin in 2001. After his Habilitation, he was visiting professor in theoretical as well as in practical computer science at different German Universities and was an assistant professor (Hochschuldozent) in Software Engineering at Paderborn University from 2002 to 2007.
His research interests include formal methods and their application in software and systems engineering and business process management. Currently, he is working on formalizing and unifying the concepts of business process modelling and on techniques and tools for the automatic analysis and verification of system and process models.
He is also working in the area of Model-based Software Engineering and techniques that make use of models in the software development process for getting rid of low-level programming. One of his favourite research topics are modelling techniques that allow to model the behaviour and the actual functionality of the software in such a way, that these models (and the code generated from them) integrates with existing software and other software models.