Advanced search
2 files | 1.55 MB Add to list

RPPM : rapid performance prediction of multithreaded applications on multicore hardware

Author
Organization
Abstract
This paper proposes RPPM which, based on a microarchitecture-independent profile of a multithreaded application, predicts its performance on a previously unseen multicore platform. RPPM breaks up multithreaded program execution into epochs based on synchronization primitives, and then predicts per-epoch active execution times for each thread and synchronization overhead to arrive at a prediction for overall application performance. RPPM predicts performance within 12 percent on average (27 percent max error) compared to cycle-level simulation. We present a case study to illustrate that RPPM can be used for making accurate multicore design trade-offs early in the design cycle.
Keywords
hardware and architecture, modeling, micro-architecture, performance, multi-threaded

Downloads

  • (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 680.91 KB
  • 8585865 accepted.pdf
    • full text (Accepted manuscript)
    • |
    • open access
    • |
    • PDF
    • |
    • 870.03 KB

Citation

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

MLA
De Pestel, Sander, et al. “RPPM : Rapid Performance Prediction of Multithreaded Applications on Multicore Hardware.” IEEE COMPUTER ARCHITECTURE LETTERS, vol. 17, no. 2, 2018, pp. 183–86, doi:10.1109/lca.2018.2849983.
APA
De Pestel, S., Van den Steen, S., Akram, S., & Eeckhout, L. (2018). RPPM : rapid performance prediction of multithreaded applications on multicore hardware. IEEE COMPUTER ARCHITECTURE LETTERS, 17(2), 183–186. https://doi.org/10.1109/lca.2018.2849983
Chicago author-date
De Pestel, Sander, Sam Van den Steen, Shoaib Akram, and Lieven Eeckhout. 2018. “RPPM : Rapid Performance Prediction of Multithreaded Applications on Multicore Hardware.” IEEE COMPUTER ARCHITECTURE LETTERS 17 (2): 183–86. https://doi.org/10.1109/lca.2018.2849983.
Chicago author-date (all authors)
De Pestel, Sander, Sam Van den Steen, Shoaib Akram, and Lieven Eeckhout. 2018. “RPPM : Rapid Performance Prediction of Multithreaded Applications on Multicore Hardware.” IEEE COMPUTER ARCHITECTURE LETTERS 17 (2): 183–186. doi:10.1109/lca.2018.2849983.
Vancouver
1.
De Pestel S, Van den Steen S, Akram S, Eeckhout L. RPPM : rapid performance prediction of multithreaded applications on multicore hardware. IEEE COMPUTER ARCHITECTURE LETTERS. 2018;17(2):183–6.
IEEE
[1]
S. De Pestel, S. Van den Steen, S. Akram, and L. Eeckhout, “RPPM : rapid performance prediction of multithreaded applications on multicore hardware,” IEEE COMPUTER ARCHITECTURE LETTERS, vol. 17, no. 2, pp. 183–186, 2018.
@article{8585865,
  abstract     = {{This paper proposes RPPM which, based on a microarchitecture-independent profile of a multithreaded application, predicts its performance on a previously unseen multicore platform. RPPM breaks up multithreaded program execution into epochs based on synchronization primitives, and then predicts per-epoch active execution times for each thread and synchronization overhead to arrive at a prediction for overall application performance. RPPM predicts performance within 12 percent on average (27 percent max error) compared to cycle-level simulation. We present a case study to illustrate that RPPM can be used for making accurate multicore design trade-offs early in the design cycle.}},
  author       = {{De Pestel, Sander and Van den Steen, Sam and Akram, Shoaib and Eeckhout, Lieven}},
  issn         = {{1556-6056}},
  journal      = {{IEEE COMPUTER ARCHITECTURE LETTERS}},
  keywords     = {{hardware and architecture,modeling,micro-architecture,performance,multi-threaded}},
  language     = {{eng}},
  number       = {{2}},
  pages        = {{183--186}},
  title        = {{RPPM : rapid performance prediction of multithreaded applications on multicore hardware}},
  url          = {{http://dx.doi.org/10.1109/lca.2018.2849983}},
  volume       = {{17}},
  year         = {{2018}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: