Advanced search
1 file | 1.83 MB Add to list

DEP+BURST : online DVFS performance prediction for energy-efficient managed language execution

Shoaib Akram (UGent) , Jennifer Sartor (UGent) and Lieven Eeckhout (UGent)
(2017) IEEE TRANSACTIONS ON COMPUTERS. 66(4). p.601-615
Author
Organization
Abstract
Making modern computer systems energy-efficient is of paramount importance. Dynamic Voltage and Frequency Scaling (DVFS) is widely used to manage the energy and power consumption in modern processors; however, for DVFS to be effective, we need the ability to accurately predict the performance impact of scaling a processor's voltage and frequency. No accurate performance predictors exist for multithreaded applications, let alone managed language applications. In this work, we propose DEP+BURST, a new performance predictor for managed multithreaded applications that takes into account synchronization, inter-thread dependencies, and store bursts, which frequently occur in managed language workloads. Our predictor lowers the performance estimation error from 27 percent for a state-of-the-art predictor to 6 percent on average, for a set of multithreaded Java applications when the frequency is scaled from 1 to 4 GHz. We also novelly propose an energy management framework that uses DEP+BURST to reduce energy consumption. We first target reducing the processor's energy consumption by lowering its frequency and hence its power consumption, while staying within a user-specified maximum slowdown threshold. For a slowdown of 5 and 10 percent, our energy manager reduces on average 13 and 19 percent of energy consumed by the memory-intensive benchmarks. We then use the energy manager to optimize total system energy, achieving an average reduction of 15.6 percent for a set of Java benchmarks. Accurate performance predictors are key to achieving high performance while keeping energy consumption low for managed language applications using DVFS.
Keywords
Energy consumption, Instruction sets, Java, Load modeling, Synchronization, Time-frequency analysis, Dynamic voltage and frequency scaling, dynamic energy management, managed runtimes, multithreaded performance estimation, POWER

Downloads

  • (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.83 MB

Citation

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

MLA
Akram, Shoaib, et al. “DEP+BURST : Online DVFS Performance Prediction for Energy-Efficient Managed Language Execution.” IEEE TRANSACTIONS ON COMPUTERS, vol. 66, no. 4, 2017, pp. 601–15, doi:10.1109/TC.2016.2609903.
APA
Akram, S., Sartor, J., & Eeckhout, L. (2017). DEP+BURST : online DVFS performance prediction for energy-efficient managed language execution. IEEE TRANSACTIONS ON COMPUTERS, 66(4), 601–615. https://doi.org/10.1109/TC.2016.2609903
Chicago author-date
Akram, Shoaib, Jennifer Sartor, and Lieven Eeckhout. 2017. “DEP+BURST : Online DVFS Performance Prediction for Energy-Efficient Managed Language Execution.” IEEE TRANSACTIONS ON COMPUTERS 66 (4): 601–15. https://doi.org/10.1109/TC.2016.2609903.
Chicago author-date (all authors)
Akram, Shoaib, Jennifer Sartor, and Lieven Eeckhout. 2017. “DEP+BURST : Online DVFS Performance Prediction for Energy-Efficient Managed Language Execution.” IEEE TRANSACTIONS ON COMPUTERS 66 (4): 601–615. doi:10.1109/TC.2016.2609903.
Vancouver
1.
Akram S, Sartor J, Eeckhout L. DEP+BURST : online DVFS performance prediction for energy-efficient managed language execution. IEEE TRANSACTIONS ON COMPUTERS. 2017;66(4):601–15.
IEEE
[1]
S. Akram, J. Sartor, and L. Eeckhout, “DEP+BURST : online DVFS performance prediction for energy-efficient managed language execution,” IEEE TRANSACTIONS ON COMPUTERS, vol. 66, no. 4, pp. 601–615, 2017.
@article{8514659,
  abstract     = {Making modern computer systems energy-efficient is of paramount importance. Dynamic Voltage and Frequency Scaling (DVFS) is widely used to manage the energy and power consumption in modern processors; however, for DVFS to be effective, we need the ability to accurately predict the performance impact of scaling a processor's voltage and frequency. No accurate performance predictors exist for multithreaded applications, let alone managed language applications. In this work, we propose DEP+BURST, a new performance predictor for managed multithreaded applications that takes into account synchronization, inter-thread dependencies, and store bursts, which frequently occur in managed language workloads. Our predictor lowers the performance estimation error from 27 percent for a state-of-the-art predictor to 6 percent on average, for a set of multithreaded Java applications when the frequency is scaled from 1 to 4 GHz. We also novelly propose an energy management framework that uses DEP+BURST to reduce energy consumption. We first target reducing the processor's energy consumption by lowering its frequency and hence its power consumption, while staying within a user-specified maximum slowdown threshold. For a slowdown of 5 and 10 percent, our energy manager reduces on average 13 and 19 percent of energy consumed by the memory-intensive benchmarks. We then use the energy manager to optimize total system energy, achieving an average reduction of 15.6 percent for a set of Java benchmarks. Accurate performance predictors are key to achieving high performance while keeping energy consumption low for managed language applications using DVFS.},
  author       = {Akram, Shoaib and Sartor, Jennifer and Eeckhout, Lieven},
  issn         = {0018-9340},
  journal      = {IEEE TRANSACTIONS ON COMPUTERS},
  keywords     = {Energy consumption,Instruction sets,Java,Load modeling,Synchronization,Time-frequency analysis,Dynamic voltage and frequency scaling,dynamic energy management,managed runtimes,multithreaded performance estimation,POWER},
  language     = {eng},
  number       = {4},
  pages        = {601--615},
  title        = {DEP+BURST : online DVFS performance prediction for energy-efficient managed language execution},
  url          = {http://dx.doi.org/10.1109/TC.2016.2609903},
  volume       = {66},
  year         = {2017},
}

Altmetric
View in Altmetric
Web of Science
Times cited: