Advanced search
1 file | 903.60 KB

A mechanistic performance model for superscalar in-order processors

Maximilien Breughe (UGent) , Stijn Eyerman (UGent) and Lieven Eeckhout (UGent)
Author
Organization
Abstract
Mechanistic processor performance modeling builds an analytical model from understanding the underlying mechanisms in the processor and provides fundamental insight in program-microarchitecture interactions, as well as microarchitecture structure scaling trends and interactions. Whereas prior work in mechanistic performance modeling focused on superscalar out-of-order processors, this paper presents a mechanistic performance model for superscalar in-order processors. We find mechanistic modeling for in-order processors to be more challenging compared to out-of-order processors because the latter are designed to hide latencies, and hence from a modeling perspective, detailed modeling of instruction execution latencies and dependencies is not required. The proposed mechanistic performance model for superscalar in-order processors models the impact of non-unit instruction execution latencies, inter-instruction dependencies, cache/TLB misses and branch mispredictions, and achieves an average performance prediction error of 2.5% compared to detailed cycle-accurate simulation. We extensively evaluate the model’s accuracy and we demonstrate its usefulness through three applications: (i) we compare in-order versus out-of-order performance, (ii) we quantify the impact of compiler optimizations on in-order performance, and (iii) we perform a power/performance design space exploration.
Keywords
micro architecture, resource scaling, embedded system, compiler optimization, cpi stack, microprocessor, mechanistic model, computer architecture, design space exploration, superscalar in-order processor, performance

Downloads

  • paper.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 903.60 KB

Citation

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

Chicago
Breughe, Maximilien, Stijn Eyerman, and Lieven Eeckhout. 2012. “A Mechanistic Performance Model for Superscalar In-order Processors.” In IEEE International Symposium on Performance Analysis of Systems and Software, Proceedings, 14–24. Piscataway, NJ, USA: IEEE.
APA
Breughe, M., Eyerman, S., & Eeckhout, L. (2012). A mechanistic performance model for superscalar in-order processors. IEEE international symposium on performance analysis of systems and software, Proceedings (pp. 14–24). Presented at the IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS - 2012), Piscataway, NJ, USA: IEEE.
Vancouver
1.
Breughe M, Eyerman S, Eeckhout L. A mechanistic performance model for superscalar in-order processors. IEEE international symposium on performance analysis of systems and software, Proceedings. Piscataway, NJ, USA: IEEE; 2012. p. 14–24.
MLA
Breughe, Maximilien, Stijn Eyerman, and Lieven Eeckhout. “A Mechanistic Performance Model for Superscalar In-order Processors.” IEEE International Symposium on Performance Analysis of Systems and Software, Proceedings. Piscataway, NJ, USA: IEEE, 2012. 14–24. Print.
@inproceedings{2093763,
  abstract     = {Mechanistic processor performance modeling builds an analytical model from understanding the underlying mechanisms in the processor and provides fundamental insight in program-microarchitecture interactions, as well as microarchitecture structure scaling trends and interactions. Whereas prior work in mechanistic performance modeling focused on superscalar out-of-order processors, this paper presents a mechanistic performance model for superscalar in-order processors. We find mechanistic modeling for in-order processors to be more challenging compared to out-of-order processors because the latter are designed to hide latencies, and hence from a modeling perspective, detailed modeling of instruction execution latencies and dependencies is not required. The proposed mechanistic performance model for superscalar in-order processors models the impact of non-unit instruction execution latencies, inter-instruction dependencies, cache/TLB misses and branch mispredictions, and achieves an average performance prediction error of 2.5\% compared to detailed cycle-accurate simulation. We extensively evaluate the model{\textquoteright}s accuracy and we demonstrate its usefulness through three applications: (i) we compare in-order versus out-of-order performance, (ii) we quantify the impact of compiler optimizations on in-order performance, and (iii) we perform a power/performance design space exploration.},
  articleno    = {2},
  author       = {Breughe, Maximilien and Eyerman, Stijn and Eeckhout, Lieven},
  booktitle    = {IEEE international symposium on performance analysis of systems and software, Proceedings},
  isbn         = {9781467311441},
  keyword      = {micro architecture,resource scaling,embedded system,compiler optimization,cpi stack,microprocessor,mechanistic model,computer architecture,design space exploration,superscalar in-order processor,performance},
  language     = {eng},
  location     = {New Brunswick, NJ, USA},
  pages        = {2:14--2:24},
  publisher    = {IEEE},
  title        = {A mechanistic performance model for superscalar in-order processors},
  year         = {2012},
}