Document Type
Article
Department/Program
Computer Science
Journal Title
Acm Sigplan Notices
Pub Date
2014
Publisher
49
Issue
10
First Page
745
Abstract
Predicting a sequence of upcoming function calls is important for optimizing programs written in modern managed languages (e.g., Java, Javascript, C#.) Existing function call predictions are mainly built on statistical patterns, suitable for predicting a single call but not a sequence of calls. This paper presents a new way to enable call sequence prediction, which exploits program structures through Probabilistic Calling Automata (PCA), a new program representation that captures both the inherent ensuing relations among function calls, and the probabilistic nature of execution paths. It shows that PCA-based prediction outperforms existing predictions, yielding substantial speedup when being applied to guide Just-In-Time compilation. By enabling accurate, efficient call sequence prediction for the first time, PCA-based predictors open up many new opportunities for dynamic program optimizations.
Recommended Citation
Zhao, Z., Wu, B., Zhou, M., Ding, Y., Sun, J., Shen, X., & Wu, Y. (2014, October). Call sequence prediction through probabilistic calling automata. In ACM SIGPLAN Notices (Vol. 49, No. 10, pp. 745-762). ACM.
DOI
10.1145/2660193.2660221
