Advanced search
1 file | 52.13 KB Add to list

Using HPM-sampling to drive dynamic compilation

(2007) ACM SIGPLAN NOTICES. 42(10). p.553-567
Author
Organization
Abstract
All high-performance production JVMs employ an adaptive strategy for program execution. Methods are first executed unoptimized and then an online profiling mechanism is used to find a subset of methods that should be optimized during the same execution. This paper empirically evaluates the design space of several profilers for initiating dynamic compilation and shows that existing online profiling schemes suffer from several limitations. They provide an insufficient number of samples, are untimely, and have limited accuracy at determining the frequently executed methods. We describe and comprehensively evaluate HPM-sampling, a simple but effective profiling scheme for finding optimization candidates using hardware performance monitors (HPMs) that addresses the aforementioned limitations. We show that HPM-sampling is more accurate; has low overhead; and improves performance by 5.7% on average and up to 18.3% when compared to the default system in Jikes RVM, without changing the compiler.
Keywords
Hardware Performance Monitors, Java, Profiling}, Just-in-time compilation

Downloads

  • P107 212.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 52.13 KB

Citation

Please use this url to cite or link to this publication:

MLA
Buytaert, Dries et al. “Using HPM-sampling to Drive Dynamic Compilation.” ACM SIGPLAN NOTICES 42.10 (2007): 553–567. Print.
APA
Buytaert, D., Georges, A., Hind, M., Arnold, M., Eeckhout, L., & De Bosschere, K. (2007). Using HPM-sampling to drive dynamic compilation. ACM SIGPLAN NOTICES, 42(10), 553–567. Presented at the 22nd International Conference on Object-Oriented Programming, Systems, Languages and Applications.
Chicago author-date
Buytaert, Dries, Andy Georges, Michael Hind, Matthew Arnold, Lieven Eeckhout, and Koen De Bosschere. 2007. “Using HPM-sampling to Drive Dynamic Compilation.” Acm Sigplan Notices 42 (10): 553–567.
Chicago author-date (all authors)
Buytaert, Dries, Andy Georges, Michael Hind, Matthew Arnold, Lieven Eeckhout, and Koen De Bosschere. 2007. “Using HPM-sampling to Drive Dynamic Compilation.” Acm Sigplan Notices 42 (10): 553–567.
Vancouver
1.
Buytaert D, Georges A, Hind M, Arnold M, Eeckhout L, De Bosschere K. Using HPM-sampling to drive dynamic compilation. ACM SIGPLAN NOTICES. ACM Press; 2007;42(10):553–67.
IEEE
[1]
D. Buytaert, A. Georges, M. Hind, M. Arnold, L. Eeckhout, and K. De Bosschere, “Using HPM-sampling to drive dynamic compilation,” ACM SIGPLAN NOTICES, vol. 42, no. 10, pp. 553–567, 2007.
@article{417218,
  abstract     = {All high-performance production JVMs employ an adaptive strategy for program execution. Methods are first executed unoptimized and then an online profiling mechanism is used to find a subset of methods that should be optimized during the same execution. This paper empirically evaluates the design space of several profilers for initiating dynamic compilation and shows that existing online profiling schemes suffer from several limitations. They provide an insufficient number of samples, are untimely, and have limited accuracy at determining the frequently executed methods. We describe and comprehensively evaluate HPM-sampling, a simple but effective profiling scheme for finding optimization candidates using hardware performance monitors (HPMs) that addresses the aforementioned limitations. We show that HPM-sampling is more accurate; has low overhead; and improves performance by 5.7% on average and up to 18.3% when compared to the default system in Jikes RVM, without changing the compiler.},
  author       = {Buytaert, Dries and Georges, Andy and Hind, Michael and Arnold, Matthew and Eeckhout, Lieven and De Bosschere, Koen},
  issn         = {0362-1340},
  journal      = {ACM SIGPLAN NOTICES},
  keywords     = {Hardware Performance Monitors,Java,Profiling},Just-in-time compilation},
  language     = {eng},
  location     = {Montreal, Canada},
  number       = {10},
  pages        = {553--567},
  publisher    = {ACM Press},
  title        = {Using HPM-sampling to drive dynamic compilation},
  volume       = {42},
  year         = {2007},
}

Web of Science
Times cited: