Advanced search
1 file | 836.29 KB

SWEEP: evaluating computer system energy efficiency using synthetic workloads

Kristof Du Bois (UGent) , Tim Schaeps (UGent) , Stijn Polfliet (UGent) , Frederick Ryckbosch (UGent) and Lieven Eeckhout (UGent)
Author
Organization
Abstract
Energy efficiency is a key design concern in contemporary processor and system design, in the embedded domain as well as in the enterprise domain. The focus on energy efficiency has led to a number of power benchmarking methods recently. For example, EEMBC released EnergyBench and SPEC released SPECpower to quantify a system's energy efficiency; also academics have proposed power benchmarks, such as JouleSort. A major limitation for each of these proposals is that they are tied to a specific benchmark, and hence, they provide limited insight with respect to why one system may be more energy-efficient than another. This paper proposes SWEEP, Synthetic Workloads for Energy Efficiency and Performance evaluation, a framework for generating synthetic workloads with specific behavioral characteristics. We employ SWEEP to generate a wide range of synthetic workloads while varying the instruction mix, ILP, memory access patterns, and I/O-intensiveness; and we use SWEEP to evaluate the energy efficiency of commercial computer systems across the workload space and learn about how the energy efficiency of a computer system is tied to its workload's characteristics. This paper also presents the Energy-Delay Diagram (EDD), a novel method for visualizing energy efficiency. The EDD clearly illustrates the energy versus performance trade-off, and provides more intuitive insight than the traditionally used EDP and ED2P metrics.
Keywords
workload characterization and generation, Energy-efficiency

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 836.29 KB

Citation

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

Chicago
Du Bois, Kristof, Tim Schaeps, Stijn Polfliet, Frederick Ryckbosch, and Lieven Eeckhout. 2011. “SWEEP: Evaluating Computer System Energy Efficiency Using Synthetic Workloads.” In HiPEAC  ’11 : Proceedings of the 6th International Conference on High Performance and Embedded Architectures and Compilers, 159–166. New York, NY, USA: Association for Computing Machinery (ACM).
APA
Du Bois, K., Schaeps, T., Polfliet, S., Ryckbosch, F., & Eeckhout, L. (2011). SWEEP: evaluating computer system energy efficiency using synthetic workloads. HiPEAC  ’11 : proceedings of the 6th international conference on high performance and embedded architectures and compilers (pp. 159–166). Presented at the 6th International conference on High Performance and Embedded Architectures and Compilers (HiPEAC 2011), New York, NY, USA: Association for Computing Machinery (ACM).
Vancouver
1.
Du Bois K, Schaeps T, Polfliet S, Ryckbosch F, Eeckhout L. SWEEP: evaluating computer system energy efficiency using synthetic workloads. HiPEAC  ’11 : proceedings of the 6th international conference on high performance and embedded architectures and compilers. New York, NY, USA: Association for Computing Machinery (ACM); 2011. p. 159–66.
MLA
Du Bois, Kristof, Tim Schaeps, Stijn Polfliet, et al. “SWEEP: Evaluating Computer System Energy Efficiency Using Synthetic Workloads.” HiPEAC  ’11 : Proceedings of the 6th International Conference on High Performance and Embedded Architectures and Compilers. New York, NY, USA: Association for Computing Machinery (ACM), 2011. 159–166. Print.
@inproceedings{1195549,
  abstract     = {Energy efficiency is a key design concern in contemporary processor and system design, in the embedded domain as well as in the enterprise domain. The focus on energy efficiency has led to a number of power benchmarking methods recently. For example, EEMBC released EnergyBench and SPEC released SPECpower to quantify a system's energy efficiency; also academics have proposed power benchmarks, such as JouleSort. A major limitation for each of these proposals is that they are tied to a specific benchmark, and hence, they provide limited insight with respect to why one system may be more energy-efficient than another.
This paper proposes SWEEP, Synthetic Workloads for Energy Efficiency and Performance evaluation, a framework for generating synthetic workloads with specific behavioral characteristics. We employ SWEEP to generate a wide range of synthetic workloads while varying the instruction mix, ILP, memory access patterns, and I/O-intensiveness; and we use SWEEP to evaluate the energy efficiency of commercial computer systems across the workload space and learn about how the energy efficiency of a computer system is tied to its workload's characteristics.
This paper also presents the Energy-Delay Diagram (EDD), a novel method for visualizing energy efficiency. The EDD clearly illustrates the energy versus performance trade-off, and provides more intuitive insight than the traditionally used EDP and ED2P metrics.},
  author       = {Du Bois, Kristof and Schaeps, Tim and Polfliet, Stijn and Ryckbosch, Frederick and Eeckhout, Lieven},
  booktitle    = {HiPEAC '11 : proceedings of the 6th international conference on high performance and embedded architectures and compilers},
  isbn         = {9781450302418},
  keyword      = {workload characterization and generation,Energy-efficiency},
  language     = {eng},
  location     = {Heraklion, Crete, Greece},
  pages        = {159--166},
  publisher    = {Association for Computing Machinery (ACM)},
  title        = {SWEEP: evaluating computer system energy efficiency using synthetic workloads},
  url          = {http://dx.doi.org/10.1145/1944862.1944886},
  year         = {2011},
}

Altmetric
View in Altmetric