Energy-efficient dynamic task scheduling algorithms for DVS systems

Research output: Contribution to journalArticle

77 Citations (Scopus)

Abstract

Dynamic voltage scaling (DVS) is a well-known low-power design technique that reduces the processor energy by slowing down the DVS processor and stretching the task execution time. However, in a DVS system consisting of a DVS processor and multiple devices, slowing down the processor increases the device energy consumption and thereby the system-level energy consumption. In this paper, we first use system-level energy consideration to derive the optimal scaling factor by which a task should be scaled if there are no deadline constraints. Next, we develop dynamic task-scheduling algorithms that make use of dynamic processor utilization and optimal scaling factor to determine the speed setting of a task. We present algorithm duEDF, which reduces the CPU energy consumption and algorithm duSYS and its reduced preemption version, duSYS_PC, which reduce the system-level energy. Experimental results on the video-phone task set show that when the CPU power is dominant, algorithm duEDF results in up to 45 energy savings compared to the non-DVS case. When the CPU power and device power are comparable, algorithms duSYS and duSYS_PC achieve up to 25 energy saving compared to CPU energy-efficient algorithm duEDF, and up to 12 energy saving over the non-DVS scheduling algorithm. However, if the device power is large compared to the CPU power, then we show that a DVS scheme does not result in lowest energy. Finally, a comparison of the performance of algorithms duSYS and duSYS_PC show that preemption control has minimal effect on system-level energy reduction.

Original languageEnglish (US)
Article number17
JournalTransactions on Embedded Computing Systems
Volume7
Issue number2
DOIs
StatePublished - Feb 1 2008

Fingerprint

Scheduling algorithms
Program processors
Electron energy levels
Energy conservation
Energy utilization
Voltage scaling
Stretching

Keywords

  • DVS system
  • Dynamic task scheduling
  • Energy minimization
  • Optimal scaling factor
  • Real time

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software

Cite this

Energy-efficient dynamic task scheduling algorithms for DVS systems. / Zhuo, Jianli; Chakrabarti, Chaitali.

In: Transactions on Embedded Computing Systems, Vol. 7, No. 2, 17, 01.02.2008.

Research output: Contribution to journalArticle

@article{7251beff8f9447b5ae5f6f1df703de37,
title = "Energy-efficient dynamic task scheduling algorithms for DVS systems",
abstract = "Dynamic voltage scaling (DVS) is a well-known low-power design technique that reduces the processor energy by slowing down the DVS processor and stretching the task execution time. However, in a DVS system consisting of a DVS processor and multiple devices, slowing down the processor increases the device energy consumption and thereby the system-level energy consumption. In this paper, we first use system-level energy consideration to derive the optimal scaling factor by which a task should be scaled if there are no deadline constraints. Next, we develop dynamic task-scheduling algorithms that make use of dynamic processor utilization and optimal scaling factor to determine the speed setting of a task. We present algorithm duEDF, which reduces the CPU energy consumption and algorithm duSYS and its reduced preemption version, duSYS_PC, which reduce the system-level energy. Experimental results on the video-phone task set show that when the CPU power is dominant, algorithm duEDF results in up to 45 energy savings compared to the non-DVS case. When the CPU power and device power are comparable, algorithms duSYS and duSYS_PC achieve up to 25 energy saving compared to CPU energy-efficient algorithm duEDF, and up to 12 energy saving over the non-DVS scheduling algorithm. However, if the device power is large compared to the CPU power, then we show that a DVS scheme does not result in lowest energy. Finally, a comparison of the performance of algorithms duSYS and duSYS_PC show that preemption control has minimal effect on system-level energy reduction.",
keywords = "DVS system, Dynamic task scheduling, Energy minimization, Optimal scaling factor, Real time",
author = "Jianli Zhuo and Chaitali Chakrabarti",
year = "2008",
month = "2",
day = "1",
doi = "10.1145/1331331.1331341",
language = "English (US)",
volume = "7",
journal = "ACM Transactions on Embedded Computing Systems",
issn = "1539-9087",
publisher = "Association for Computing Machinery (ACM)",
number = "2",

}

TY - JOUR

T1 - Energy-efficient dynamic task scheduling algorithms for DVS systems

AU - Zhuo, Jianli

AU - Chakrabarti, Chaitali

PY - 2008/2/1

Y1 - 2008/2/1

N2 - Dynamic voltage scaling (DVS) is a well-known low-power design technique that reduces the processor energy by slowing down the DVS processor and stretching the task execution time. However, in a DVS system consisting of a DVS processor and multiple devices, slowing down the processor increases the device energy consumption and thereby the system-level energy consumption. In this paper, we first use system-level energy consideration to derive the optimal scaling factor by which a task should be scaled if there are no deadline constraints. Next, we develop dynamic task-scheduling algorithms that make use of dynamic processor utilization and optimal scaling factor to determine the speed setting of a task. We present algorithm duEDF, which reduces the CPU energy consumption and algorithm duSYS and its reduced preemption version, duSYS_PC, which reduce the system-level energy. Experimental results on the video-phone task set show that when the CPU power is dominant, algorithm duEDF results in up to 45 energy savings compared to the non-DVS case. When the CPU power and device power are comparable, algorithms duSYS and duSYS_PC achieve up to 25 energy saving compared to CPU energy-efficient algorithm duEDF, and up to 12 energy saving over the non-DVS scheduling algorithm. However, if the device power is large compared to the CPU power, then we show that a DVS scheme does not result in lowest energy. Finally, a comparison of the performance of algorithms duSYS and duSYS_PC show that preemption control has minimal effect on system-level energy reduction.

AB - Dynamic voltage scaling (DVS) is a well-known low-power design technique that reduces the processor energy by slowing down the DVS processor and stretching the task execution time. However, in a DVS system consisting of a DVS processor and multiple devices, slowing down the processor increases the device energy consumption and thereby the system-level energy consumption. In this paper, we first use system-level energy consideration to derive the optimal scaling factor by which a task should be scaled if there are no deadline constraints. Next, we develop dynamic task-scheduling algorithms that make use of dynamic processor utilization and optimal scaling factor to determine the speed setting of a task. We present algorithm duEDF, which reduces the CPU energy consumption and algorithm duSYS and its reduced preemption version, duSYS_PC, which reduce the system-level energy. Experimental results on the video-phone task set show that when the CPU power is dominant, algorithm duEDF results in up to 45 energy savings compared to the non-DVS case. When the CPU power and device power are comparable, algorithms duSYS and duSYS_PC achieve up to 25 energy saving compared to CPU energy-efficient algorithm duEDF, and up to 12 energy saving over the non-DVS scheduling algorithm. However, if the device power is large compared to the CPU power, then we show that a DVS scheme does not result in lowest energy. Finally, a comparison of the performance of algorithms duSYS and duSYS_PC show that preemption control has minimal effect on system-level energy reduction.

KW - DVS system

KW - Dynamic task scheduling

KW - Energy minimization

KW - Optimal scaling factor

KW - Real time

UR - http://www.scopus.com/inward/record.url?scp=40549085022&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=40549085022&partnerID=8YFLogxK

U2 - 10.1145/1331331.1331341

DO - 10.1145/1331331.1331341

M3 - Article

AN - SCOPUS:40549085022

VL - 7

JO - ACM Transactions on Embedded Computing Systems

JF - ACM Transactions on Embedded Computing Systems

SN - 1539-9087

IS - 2

M1 - 17

ER -