| Ondrej Lhoták
Associate Professor Joined School 2006 BMath (Waterloo),
|
Professor Lhoták's research interests are generally in the area of programming languages, compilers, and program analysis, with a specific focus on object-oriented languages.
Program analysis has long been used to generate efficient code, and it is increasingly being used in software engineering tools. These applications require precise and efficient program analyses. The increased modularity enabled by object-oriented languages makes interprocedural analysis necessary for precise results. Therefore, Professor Lhoták is working on making precise interprocedural analyses efficient enough to be practical.
Specifically, Professor Lhoták and colleagues explored the use of binary decision diagrams (BDDs) to represent the large data structures that must be manipulated in interprocedural analyses. In particular, BDDs have been shown to be able to compactly encode context-sensitive analysis facts. Currently, Professor Lhoták is using his BDD-based analysis implementation to study the effects of context sensitivity on analysis precision, and to design efficient context-sensitive analyses.
Professor Lhoták and colleagues have applied program analyses both to generating efficient code (for Java and AspectJ), and to aid programmers through software development tools (in particular, the Eclipse IDE).
Early Researcher Award (2010-2015); NSERC Discovery Accelerator Supplement (2010-2013); IBM Center for Advanced Studies (CAS) Faculty Fellow (2008)
Professor Lhoták is one of the developers of abc, one of the two leading compilers for the AspectJ language.
Prior to graduate school, Professor Lhoták worked at iAnywhere/Sybase in Waterloo on the query execution engine of SQL Anywhere.
Naeem N., Lhoták O., Rodriguez J., Practical Extensions to the IFDS Algorithm, CC 2010, March 2010.
Lhoták O., Curial S., Amaral J. N., Using XBDDs and ZBDDs in points-to analysis, SP&E 39(2):163-188, 2009.
Lhoták O., Hendren L., Relations as an abstraction for BDD-based program analysis, TOPLAS 30, 4, Article 19, 2008.
Naeem, N., Lhoták O., Typestate-like analysis of multiple interacting objects, OOPSLA 2008, October 2008.
Lhoták O., Hendren L., Evaluating the benefits of context-sensitive points-to analysis using a BDD-based implementation, TOSEM 18, 1, Article 3, 2008.
M. Berndl, O. Lhoták, F. Qian, L. Hendren, and N. Umanee. Points-to Analysis using BDDs. Proceedings of the ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI), pp. 103-114, 2003.
O. Lhoták and L. Hendren. Scaling Java Points-to Analysis using Spark. Proceedings of the International Conference on Compiler Construction (CC), pp. 170-184, 2003.

David R. Cheriton School of Computer Science
University of Waterloo
Waterloo, Ontario, Canada N2L 3G1
Tel: 519-888-4567 x33293
Fax: 519-885-1208
Contact | Feedback: cs-webmaster@cs.uwaterloo.ca | David R. Cheriton School of Computer Science | Faculty of Mathematics