Dr. Victor Pankratius, MIT Computer Science and Artificial Intelligence Laboratory (CSAIL)
Title: Improving Multicore Software Quality With Data Mining and Crowdsourcing Techniques
Abstract: Recent empirical studies have shown the need to extend the capabilities of standard debugging tools to detect a larger spectrum of parallel programming errors. This talk contrasts debugging effort and problems typically encountered by programmers in Java, Scala, OpenMP, and C with Pthreads and Transactional Memory extensions. Learning from the results of these studies, two new directions are promising to achieve better multicore software quality. The talk shows how data mining can help identify a broad range of errors based on the analysis of anomalous program behavior. Moreover, language usability engineering emerges as an important field to improve parallel software in the long run. Debugging can often be avoided altogether if programming languages provide parallel constructs with reduced error potential and more intuitive use. The automated collection and analysis of data from code repositories and instrumented IDEs thus becomes a key building block in a regular quality assurance process. The talk discusses a crowdsourcing approach towards a worldwide distributed debugger that learns about problems and language usability issues while millions of programmers are working on their everyday projects.
Biography: Dr. Pankratius is a visiting scientist at the Massachusetts Institute of Technology/USA in the Computer Science and Artificial Intelligence Laboratory (CSAIL). He also heads the Multicore Software Engineering investigator group at the Karlsruhe Institute of Technology, Germany. His research concentrates on how to make parallel programming easier and covers a range of research topics including auto-tuning, language design, debugging, software engineering in the cloud, and empirical studies.
Dr. Pankratius was also a visiting scientist at Intel and Sun Labs/Oracle in the US. He received several awards for his work, including the Intel Leadership Award. He is the elected chairman of the Software Engineering for Parallel Systems (SEPARS) international working group and served as a PC member and reviewer in international conferences and journals on software engineering and parallel computing. Dr. Pankratius co-chaired the Multicore Software Engineering workshop series co-located with ICSE and was a general chair of the International Conference on Multicore Software Engineering, Performance, and Tools that emerged from these workshops. He is a member of the ACM, IEEE, HiPEAC, and the German Computer Science Society.
Dr. Pankratius holds a Habilitation degree (PD) in Computer Science from the Karlsruhe Institute of Technology (2012) and a Dr.rer.pol. degree with distinction from the University of Karlsruhe (2007). From the University of Münster, Germany, he received a Diplom degree (M.S.) in Business Computer Science best of class (2003) and a Bachelor of Science in Information Systems (BScIS, 2002).