TITLE: Managing Software Complexity through Compositional Analysis

ABSTRACT:

Today’s computer systems are large, they are complex, and they are everywhere. Our lives are increasingly dependent on the correct operation of the software running them. Yet our ability to analyze and understand them seems to be lagging behind. In this talk, I will argue that the key to dealing with the complexity in software systems is through abstraction and compositional techniques. I will then present two approaches that automatically decompose software artifacts and derive the right level of abstraction for specific analysis tasks. First, I will describe an approach called FHistorian, which identifies software features and extracts feature models from version histories to enable modular software development. Then, I will present a technique, uLTR, which dissects and analyzes timing requirements for real-time component-based systems to facilitate better design and more efficient runtime monitoring. Finally, I will share some future directions for building a modular software mining and construction framework to exploit the increasing volume of existing software artifacts available on the Internet.

BIO:

Yi Li is a Ph.D. candidate in the Department of Computer Science at the University of Toronto. Previously, he received his B.Comp. degree with First Class Honours from the National University of Singapore in 2011 and his M.Sc. degree in Computer Science from the University of Toronto in 2013. His research interests include program analysis, software verification, software requirements, and history analysis. His research also addressed important problems in SMT solving techniques and artificial intelligence. His recent work on software history analysis won an ACM Distinguished Paper Award at the 30th International Conference on Automated Software Engineering (ASE’15).