Energy-aware scheduling for frame-based tasks on heterogeneous multiprocessor platforms

Dawei Li, Jie Wu

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

21 Citations (Scopus)

Abstract

Modern computational systems have adopted heterogeneous multiprocessors to increase their computation capability. As the performance increases, the energy consumption in these systems also increases significantly. Dynamic Voltage and Frequency Scaling (DVFS), which allows processors to dynamically adjust their supply voltages and execution frequencies to work on different power/energy levels, is considered an efficient scheme to achieve the goal of saving energy. In this paper, we consider scheduling frame-based tasks on DVFS-enabled heterogeneous multiprocessor platforms with the goal of achieving minimal overall energy consumption. We consider three types of heterogeneous platforms, namely, dependent platforms without runtime adjusting, dependent platforms with runtime adjusting, and independent platforms. For all of these three platforms, we first introduce a Relaxation-based Naive Rounding Algorithm (RNRA), which can produce good solutions for some cases, but may be unstable under other situations. Then, we propose a Relaxation-based Iterative Rounding Algorithm (RIRA). Experiments and comparisons show that our RIRA produces a better performance than RNRA and other existing methods, and achieves near-optimal scheduling under most cases.

Original languageEnglish
Title of host publicationProceedings - 41st International Conference on Parallel Processing, ICPP 2012
Pages430-439
Number of pages10
DOIs
StatePublished - 20 Dec 2012
Event41st International Conference on Parallel Processing, ICPP 2012 - Pittsburgh, PA, United States
Duration: 10 Sep 201213 Sep 2012

Publication series

NameProceedings of the International Conference on Parallel Processing
ISSN (Print)0190-3918

Other

Other41st International Conference on Parallel Processing, ICPP 2012
CountryUnited States
CityPittsburgh, PA
Period10/09/1213/09/12

Fingerprint

Rounding
Multiprocessor
Scheduling
Voltage
Energy
Energy Consumption
Energy utilization
Scaling
Minimal Energy
Optimal Scheduling
Dependent
Energy Saving
Energy Levels
Electron energy levels
Energy conservation
Unstable
Electric potential
Experiment
Experiments
Voltage scaling

Keywords

  • Heterogeneous multiprocessor platforms
  • dynamic voltage and frequency scaling (DVFS)
  • energy-aware scheduling
  • task partitioning

Cite this

Li, D., & Wu, J. (2012). Energy-aware scheduling for frame-based tasks on heterogeneous multiprocessor platforms. In Proceedings - 41st International Conference on Parallel Processing, ICPP 2012 (pp. 430-439). [6337604] (Proceedings of the International Conference on Parallel Processing). https://doi.org/10.1109/ICPP.2012.26
Li, Dawei ; Wu, Jie. / Energy-aware scheduling for frame-based tasks on heterogeneous multiprocessor platforms. Proceedings - 41st International Conference on Parallel Processing, ICPP 2012. 2012. pp. 430-439 (Proceedings of the International Conference on Parallel Processing).
@inproceedings{e42045b9ee4d4a959a2ff9ceb1b3c95c,
title = "Energy-aware scheduling for frame-based tasks on heterogeneous multiprocessor platforms",
abstract = "Modern computational systems have adopted heterogeneous multiprocessors to increase their computation capability. As the performance increases, the energy consumption in these systems also increases significantly. Dynamic Voltage and Frequency Scaling (DVFS), which allows processors to dynamically adjust their supply voltages and execution frequencies to work on different power/energy levels, is considered an efficient scheme to achieve the goal of saving energy. In this paper, we consider scheduling frame-based tasks on DVFS-enabled heterogeneous multiprocessor platforms with the goal of achieving minimal overall energy consumption. We consider three types of heterogeneous platforms, namely, dependent platforms without runtime adjusting, dependent platforms with runtime adjusting, and independent platforms. For all of these three platforms, we first introduce a Relaxation-based Naive Rounding Algorithm (RNRA), which can produce good solutions for some cases, but may be unstable under other situations. Then, we propose a Relaxation-based Iterative Rounding Algorithm (RIRA). Experiments and comparisons show that our RIRA produces a better performance than RNRA and other existing methods, and achieves near-optimal scheduling under most cases.",
keywords = "Heterogeneous multiprocessor platforms, dynamic voltage and frequency scaling (DVFS), energy-aware scheduling, task partitioning",
author = "Dawei Li and Jie Wu",
year = "2012",
month = "12",
day = "20",
doi = "10.1109/ICPP.2012.26",
language = "English",
isbn = "9780769547961",
series = "Proceedings of the International Conference on Parallel Processing",
pages = "430--439",
booktitle = "Proceedings - 41st International Conference on Parallel Processing, ICPP 2012",

}

Li, D & Wu, J 2012, Energy-aware scheduling for frame-based tasks on heterogeneous multiprocessor platforms. in Proceedings - 41st International Conference on Parallel Processing, ICPP 2012., 6337604, Proceedings of the International Conference on Parallel Processing, pp. 430-439, 41st International Conference on Parallel Processing, ICPP 2012, Pittsburgh, PA, United States, 10/09/12. https://doi.org/10.1109/ICPP.2012.26

Energy-aware scheduling for frame-based tasks on heterogeneous multiprocessor platforms. / Li, Dawei; Wu, Jie.

Proceedings - 41st International Conference on Parallel Processing, ICPP 2012. 2012. p. 430-439 6337604 (Proceedings of the International Conference on Parallel Processing).

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - Energy-aware scheduling for frame-based tasks on heterogeneous multiprocessor platforms

AU - Li, Dawei

AU - Wu, Jie

PY - 2012/12/20

Y1 - 2012/12/20

N2 - Modern computational systems have adopted heterogeneous multiprocessors to increase their computation capability. As the performance increases, the energy consumption in these systems also increases significantly. Dynamic Voltage and Frequency Scaling (DVFS), which allows processors to dynamically adjust their supply voltages and execution frequencies to work on different power/energy levels, is considered an efficient scheme to achieve the goal of saving energy. In this paper, we consider scheduling frame-based tasks on DVFS-enabled heterogeneous multiprocessor platforms with the goal of achieving minimal overall energy consumption. We consider three types of heterogeneous platforms, namely, dependent platforms without runtime adjusting, dependent platforms with runtime adjusting, and independent platforms. For all of these three platforms, we first introduce a Relaxation-based Naive Rounding Algorithm (RNRA), which can produce good solutions for some cases, but may be unstable under other situations. Then, we propose a Relaxation-based Iterative Rounding Algorithm (RIRA). Experiments and comparisons show that our RIRA produces a better performance than RNRA and other existing methods, and achieves near-optimal scheduling under most cases.

AB - Modern computational systems have adopted heterogeneous multiprocessors to increase their computation capability. As the performance increases, the energy consumption in these systems also increases significantly. Dynamic Voltage and Frequency Scaling (DVFS), which allows processors to dynamically adjust their supply voltages and execution frequencies to work on different power/energy levels, is considered an efficient scheme to achieve the goal of saving energy. In this paper, we consider scheduling frame-based tasks on DVFS-enabled heterogeneous multiprocessor platforms with the goal of achieving minimal overall energy consumption. We consider three types of heterogeneous platforms, namely, dependent platforms without runtime adjusting, dependent platforms with runtime adjusting, and independent platforms. For all of these three platforms, we first introduce a Relaxation-based Naive Rounding Algorithm (RNRA), which can produce good solutions for some cases, but may be unstable under other situations. Then, we propose a Relaxation-based Iterative Rounding Algorithm (RIRA). Experiments and comparisons show that our RIRA produces a better performance than RNRA and other existing methods, and achieves near-optimal scheduling under most cases.

KW - Heterogeneous multiprocessor platforms

KW - dynamic voltage and frequency scaling (DVFS)

KW - energy-aware scheduling

KW - task partitioning

UR - http://www.scopus.com/inward/record.url?scp=84871172338&partnerID=8YFLogxK

U2 - 10.1109/ICPP.2012.26

DO - 10.1109/ICPP.2012.26

M3 - Conference contribution

SN - 9780769547961

T3 - Proceedings of the International Conference on Parallel Processing

SP - 430

EP - 439

BT - Proceedings - 41st International Conference on Parallel Processing, ICPP 2012

ER -

Li D, Wu J. Energy-aware scheduling for frame-based tasks on heterogeneous multiprocessor platforms. In Proceedings - 41st International Conference on Parallel Processing, ICPP 2012. 2012. p. 430-439. 6337604. (Proceedings of the International Conference on Parallel Processing). https://doi.org/10.1109/ICPP.2012.26