Abstract
Understanding a program from its execution traces is extremely difficult because a trace consists of thousands to millions of events, such as method calls, object creation and destruction, etc. Nonetheless, execution traces can provide valuable information, once abstracted from their low-level events. We propose to identify feature-level phases based on events collected from traces of the program execution. We cast our approach in an optimization problem, searching through the dynamic information provided by the program's execution traces to form a set of phases that minimizes coupling while maximizing cohesion. We applied and evaluated our search algorithms on different execution scenarios of JHotDraw and Pooka. © 2014 Springer International Publishing Switzerland.
Author supplied keywords
Cite
CITATION STYLE
Benomar, O., Sahraoui, H., & Poulin, P. (2014). Detecting program execution phases using heuristic search. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8636 LNCS, pp. 16–30). Springer Verlag. https://doi.org/10.1007/978-3-319-09940-8_2
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
