Keynotes
Wolfram Schulte, Microsoft Research, USA
Wolfram Schulte is a principal researcher and the founding manager of Microsoft's Research in Software Engineering (RiSE) team in Redmond, WA., USA. Wolfram's research concentrates on improving software development productivity by providing better methods, languages and tools for describing, developing, analyzing, testing and executing software. Wolfram jumpstarted RiSE's model-based testing and unit testing research. Lately he focused on building concurrent runtimes and verifying concurrent systems. Before joining Microsoft Research in 1999, Wolfram spent five years as an assistant professor for computer science at the University of Ulm, five years as a research and teaching assistant at the Technical University Berlin, and one year as an engineer at sd&m. Contact him at schulte@microsoft.com.
http://research.microsoft.com/en-us/people/schulte/
Title: Software engineering and testing at Microsoft : A research perspective
Abstract: Software is changing the world. But the software that drives this change is continuously growing in complexity and functionality. At the same time release cycles are shortening and customer expectations are rising. How does Microsoft guarantee trustworthy software in this challenging environment? In this talk I will shortly present Microsoft's approach to testing: its engineering life cycle, the role that testers play, and the used test tools and systems. I will then drill down into technologies (and tools) that my team championed in the last years: model-based testing with SpecExplorer, unit testing with Pex, security testing with Sage, and concurrency testing with Chess and friends. While these tools improve the effectiveness of software testing dramatically, there are still plenty of research challenges ahead, some of which I will address. Only if we tackle these successfully, we might be able to say someday that software bugs are as rare as power outages.
Ian Sommerville, St. Andrews University, Scotland
Ian Sommerville is Professor of Software Engineering at St Andrews University in Scotland. His research interests are in large-scale complex IT systems with a particular interest in socio-technical issues that affect the dependability and operation of these complex systems. He has worked with social scientists for many years in adapting and using techniques for social analysis in the study of work with a view to understanding how work, including software engineering and software testing, is really done. Recently, he has developed new interests in cloud computing and the socio-technical issues around the migration of application systems and portfolios to the cloud. Ian is also the author of a widely used textbook in software engineering, first published in 1982 with the 9th edition published in 2010.
http://www.software-engin.com/
Title: Designing for Failure: Challenges for developing and testing complex systems of systems
Abstract: In this talk, I will discuss some of the issues that affect the construction of large-scale complex systems of systems and will suggest that current approaches for developing and testing software are inappropriate for these systems. I will suggest that approaches to engineering based on reductionism do not scale and will explain why we need to shift focus from system correctness to system failure. I will discuss the need to design and test systems so that they can recover from 'normal failures'.
Walter Tichy, Karlsruhe Institute of Technology, Germany
Walter F. Tichy has been professor of Computer Science at the Karlsruhe Institute of Technology (formerly University Karlsruhe), Germany, since 1986, and was dean of the faculty of computer science from 2002 to 2004. Previously, he was senior scientist at Carnegie Group, Inc., in Pittsburgh, Pennsylvania and served six years on the faculty of Computer Science at Purdue University in West Lafayette, Indiana. His primary research interests are software engineering and parallelism. He is currently directing research on a variety of topics, including empirical software engineering, software architecture, and programming languages and tools for multi/manycore computers. He has consulted widely for industry. He earned an M.S. and a PhD in Computer Science from Carnegie Mellon University in 1976 and 1980, resp. He is director at the Forschungszentrum Informatik, a technology transfer institute in Karlsruhe. He is co-founder of ParTec, a company specializing in cluster computing. He has been involved in organizing numerous conferences and workshops; among others, he was program co-chair for the 25th International Conference on Software Engineering (2003). He received the Intel Award for the Advancement of Parallel Computing in 2009. Dr. Tichy is a member of ACM, GI, and the IEEE Comptuer Society.
http://www.ipd.uni-karlsruhe.de/Tichy/people.php?id=15
Title: Tunable Architectures or How to Get the Most out of Your Multicore
Abstract:
Laptops with four cores, PCs with 16, and servers 48 cores per chip -- the era of multicore has arrived. Programming these machines, however, poses many challenges. In this talk, I discuss the issue of performance tuning. Because multicore computers differ significantly in hardware characteristics, Software developers hand-tune parallel programs for a given platform. This is tedious and leads to non-portable code. Although the architecture of parallel applications also requires adaptation to achieve best performance, it is rarely modified because of the additional effort. The Tunable Architectures approach proposed in this talk automates the architecture adaptation of parallel programs and uses an auto-tuner to find the best-performing architectural choice for a given machine. We introduce a new architecture description language based on parallel patterns. Several case studies demonstrate significant performance improvements due to automatic architecture tuning and show the applicability of the approach to industrial applications. The talk also points out several other research areas in the multicore software challenge, in particular parallel testing, finding concurrency defects, and re-engineering sequential programs for parallelization.
Bernd Leukert, SAP AG
Bernd Leukert is executive vice president of Quality Governance and Production at SAP. Since 2007 he has been responsible for development and improvement of the overall SAP quality management system, for the product life cycle, and for the Ramp-Up of all SAP products into the market. Bernd Leukert also initiated more customer orientation within SAP as well as Customer Validation, in order to increase company efficiency. He reports directly to Gerhard Oswald, chief operating officer (COO) and member of the SAP Executive Board. Under the COO board area, Bernd Leukert and his team streamline services to customers and product quality at SAP. The Quality Governance and Production organization focuses on early customer involvement to ensure that only mature and robust products are shipped to the market.
Bernd Leukert holds a master's degree in business administration, focusing on engineering and information technology, from the University of Karlsruhe, Germany. He also spent one year of his studies at the Trinity College of Dublin, Ireland
Title: Customers as Integral Part of SAP’s Quality Strategy
Abstract:
Customer focus is becoming increasingly important. This talk focuses on how SAP masters this challenge and presents the new framework of SAP’s Quality Strategy – “Delight the customer”. Alongside a number of new aspects, the three programs “Customer Proof”, “Customer Validation” and “Customer Connection” were set up to integrate SAP customers even more and even earlier in the product life cycle. This is seen as a key success factor for SAP when it comes to delivering high-quality products in a sustainable way. Another important aspect of SAP’s Quality Strategy is the approach to implement LEAN, which moves us forward to a more agile product creation process. It will also be explained how SAP manages to transfer a manufacturing focused methodology to the software industry and how its impacts on software production and our customers are seen. The combination of continuous improvement processes and the focus on people make LEAN the right production model for SAP.