The Complexity Crisis
François E. Cellier, Independent Researcher, Switzerland
Context-Aware Decision Support in Dynamic Environments - Theoretical & Technological Foundations
Alexander Smirnov, SPC RAS, Russian Federation
The DAO of Parallel Software Construction
Armin Größlinger, University of Passau, Germany
Software Engineering Research under the Lamppost
Claes Wohlin, Blekinge Institute of Technology, Sweden
The Complexity Crisis
François E. Cellier
Independent Researcher
Switzerland
http://www.inf.ethz.ch/~fcellier/
Brief Bio
François E. Cellier received his BS degree in electrical engineering in 1972, his MS degree in automatic control in 1973, and his PhD degree in technical sciences in 1979, all from the Swiss Federal Institute of Technology (ETH) Zurich. Dr. Cellier worked at the University of Arizona as professor of Electrical and Computer Engineering from 1984 until 2005. He returned to his home country of Switzerland and his alma mater in the summer of 2005.Dr. Cellier's main scientific interests concern modeling and simulation methodologies, and the design of advanced software systems for simulation, computer-aided modeling, and computer-aided design. Dr. Cellier has authored or co-authored more than 200 technical publications, and he has edited several books. He published a textbook on Continuous System Modeling in 1991 and a second textbook on Continuous System Simulation in 2006, both with Springer-Verlag, New York. He is a fellow of the Society for Modeling and Simulation International (SCS).
Abstract
Engineering systems are becoming ever more complex. The total length of the wires in the cable tree of a modern jetliner adds up to several hundred kilometers. The cable tree features many thousand connections. These systems have become so complex that it is no longer possible to completely test such a system ahead of time in all of its
potential operational modes. Simulation may offer a partial answer to the problem, but also the models of these systems are becoming ever more complex.
The talk will address a variety of issues related to these problems:
1. What demands are to be met by modeling environments to enable modelers to formulate models of complex systems reliably and within reasonable time limits?
2. What demands are to be met by simulation environments to enable them to simulate large-scale models of complex physical systems reliably, effectively, and efficiently?
3. How can models of complex systems become fault-tolerant, i.e., what demands must models meet such that simulations can recover from errors on their own when they get into trouble during simulation?
Context-Aware Decision Support in Dynamic Environments - Theoretical & Technological Foundations
Alexander Smirnov
SPC RAS
Russian Federation
Brief Bio
Alexander Smirnov is a Head of Computer Aided Integrated Systems Laboratory, St. Petersburg Federal Research Center of the Russian Academy of Sciences (SPC RAS). He received his Ph.D from St. Petersburg State University of Electrical Engineering (1984) and Dr.habil. from St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences (1994), and became a Full Professor in 1998. Also, he is a Professor and a Head of International Laboratory on Intelligent Technologies for Socio-Cyberphysical Systems, ITMO University, St. Petersburg (from 2014), and a Founder of Joint Master Program on Business Informatics between ITMO University and Rostock University (Germany). He has been involved in projects sponsored by Ford, Nokia, Festo, US DoD, European Research Programs (Information Society Technologies, Esprit, Eureka/Factory, etc.), and Russian agencies in the areas of distributed intelligent systems, ontology management, intelligent decision support systems, etc.
Abstract
Context-aware decision support is required in situations happening in dynamic, rapidly changing, and often unpredictable distributed environments. Such situations can be characterized by highly decentralized up-to-date data sets coming from various resources located in cyber-physical space. The goals of context-aware support of operational decision making are to timely provide the decisions maker with up-to-date information, to assess the relevance of information & knowledge to a decision, and to gain insight in seeking and evaluating possible decision alternatives.
The lecture addresses theoretical and technological foundations of context-aware decision support. The theoretical fundamentals are built around ontologies as a widely accepted tool for the semantic modeling of context information. They provide efficient facilities to represent application knowledge, and to make resources of the dynamic environments context-aware and interoperable.
The proposed fundamentals are supported by advanced intelligent technologies (ontology management, context management, constraint satisfaction, smart space, and decision mining). An application of these ideas is illustrated by examples of decision support systems for dynamic logistics.
The DAO of Parallel Software Construction
Armin Größlinger
University of Passau
Germany
Brief Bio
Armin Größlinger received his diploma in computer science in 2003 and his PhD in 2009, both from the University of Passau. From 2009 to 2010 he worked at INRIA Saclay (France) as a research engineer. In 2010 he returned to the University of Passau as a lecturer/post-doc in the group of Prof. Christian Lengauer. He received prizes for his dissertation (2010) and for teaching excellence (2011) from the University of Passau. His research interests comprise the polyhedron model for automatic parallelization, GPGPU computing and domain-specific optimization. Currently, his main research activity is in the project "Advanced Stencil Code Engineering" involving 5 partners and funded in a priority programme of the German Research Foundation (DFG).
Abstract
Constructing parallel software has been facing increasing challenges in recent years. Diversities in hardware from embedded multicores to supercomputers, in software with a plethora of programming models and in applications from personal entertainment to high-performance computing are making manual methods for parallel software construction less efficient and less effective. To keep pace with the mounting demands on parallel software, more systematic and sustainable approaches to its construction must be employed. This talk will review selected automated techniques which have been used to build parallel software, present current challenges which must be addressed and discuss three key aspects of future parallel software construction: domain-specific languages (D), automatic program analysis (A) and automatic optimization (O).
Software Engineering Research under the Lamppost
Claes Wohlin
Blekinge Institute of Technology
Sweden
http://www.wohlin.eu/
![]()
Brief Bio
Claes Wohlin is a professor of software engineering at Blekinge Institute of Technology, Sweden. He has previously held professor chairs at the universities in Lund and Linköping in Sweden. His research interests include empirical methods in software engineering, software metrics, software quality, requirements engineering, and global software engineering. Claes Wohlin received a PhD in Communication Systems from Lund University in 1991. He has been a visiting professor at Chalmers University of Technology in Sweden (2005-2008) and a professorial visiting fellow at University of New South Wales in Australia (2009-2011). He is Editor-in-Chief of Information and Software Technology and member of the editorial boards of Software Quality Journal and Requirements Engineering Journal. He was on the editorial board of Empirical Software Engineering: An International Journal from 2002 to 2012. He has published more than 200 papers internationally in software engineering. Dr. Wohlin is the principal author of the book Experimentation in Software Engineering published by Springer in 2012, which is based on the first version of the book published by Kluwer Academic Publishers in 2000. Claes Wohlin was the recipient of the Telenor Nordic Research Prize in 2004 for his achievements in software engineering and improvement of reliability for telecommunication systems. In 2011, he was elected as a member of the Royal Swedish Academy of Engineering Sciences. Contact him at Claes.Wohlin@bth.se or visit his website at www.wohlin.eu.
Abstract
The keynote will address some of the needs and challenges in providing industrial value and make an impact on industry developing software-intensive systems. The talk will introduce five different levels of closeness in industry-academia collaboration. A model describing the fifth level of closeness in collaboration will be described; the model is focused on long-term continuous collaboration between industry and academia. Based on experiences from using this collaborative model, a top 10 list of challenges for successful collaboration between industry and academia will be presented. Further, the talk will emphasize the necessity to conduct empirical software engineering research with industry or alternatively better understand how research conducted without industrial involvement still can be made useful for industry. The latter will be exemplified with a number of different approaches to research, where the relationship to industrial development needs to be better understood.