Analysis and evaluation of greedy thread swapping based dynamic power management for MPSoC platforms

Chirag Ravishankar, Sundaram Ananthanarayanan, Siddharth Garg, Andrew Kennings

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Thread migration (TM) is a recently proposed dynamic power management technique for heterogeneous multi-processor system-on-chip (MPSoC) platforms that eliminates the area and power overheads incurred by fine-grained dynamic voltage and frequency scaling (DVFS) based power management. In this paper, we take the first step towards formally analyzing and experimentally evaluating the use of power-aware TM for parallel data streaming applications on MPSoC platforms. From an analysis perspective, we characterize the optimal mapping of threads to cores and prove the convergence properties of a complexity effective greedy thread swapping based TM algorithm to the globally optimal solution. The proposed techniques are evaluated on a 9-core FPGA based MPSoC prototype equipped with fully-functional TM and DVFS support, and running a parallelized video encoding benchmark based on the Motion Picture Experts Group (MPEG-2) standard. Our experimental results validate the proposed theoretical analysis, and show that the proposed TM algorithm provides within 8% of the DVFS performance under the same power budget, and assuming no overheads for DVFS. Assuming voltage regulator inefficiency of 80%, the proposed TM algorithm has 9% higher performance than DVFS, again under the same total power budget.

Original languageEnglish (US)
Title of host publicationProceedings of the 13th International Symposium on Quality Electronic Design, ISQED 2012
Pages617-624
Number of pages8
DOIs
StatePublished - 2012
Event13th International Symposium on Quality Electronic Design, ISQED 2012 - Santa Clara, CA, United States
Duration: Mar 19 2012Mar 21 2012

Other

Other13th International Symposium on Quality Electronic Design, ISQED 2012
CountryUnited States
CitySanta Clara, CA
Period3/19/123/21/12

Fingerprint

Motion Picture Experts Group standards
Voltage regulators
Field programmable gate arrays (FPGA)
Power management
Voltage scaling
Dynamic frequency scaling
System-on-chip

Keywords

  • DVFS
  • FPGA
  • Multi-core
  • Power management
  • Thread migration

ASJC Scopus subject areas

  • Hardware and Architecture
  • Electrical and Electronic Engineering
  • Safety, Risk, Reliability and Quality

Cite this

Ravishankar, C., Ananthanarayanan, S., Garg, S., & Kennings, A. (2012). Analysis and evaluation of greedy thread swapping based dynamic power management for MPSoC platforms. In Proceedings of the 13th International Symposium on Quality Electronic Design, ISQED 2012 (pp. 617-624). [6187557] https://doi.org/10.1109/ISQED.2012.6187557

Analysis and evaluation of greedy thread swapping based dynamic power management for MPSoC platforms. / Ravishankar, Chirag; Ananthanarayanan, Sundaram; Garg, Siddharth; Kennings, Andrew.

Proceedings of the 13th International Symposium on Quality Electronic Design, ISQED 2012. 2012. p. 617-624 6187557.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ravishankar, C, Ananthanarayanan, S, Garg, S & Kennings, A 2012, Analysis and evaluation of greedy thread swapping based dynamic power management for MPSoC platforms. in Proceedings of the 13th International Symposium on Quality Electronic Design, ISQED 2012., 6187557, pp. 617-624, 13th International Symposium on Quality Electronic Design, ISQED 2012, Santa Clara, CA, United States, 3/19/12. https://doi.org/10.1109/ISQED.2012.6187557
Ravishankar C, Ananthanarayanan S, Garg S, Kennings A. Analysis and evaluation of greedy thread swapping based dynamic power management for MPSoC platforms. In Proceedings of the 13th International Symposium on Quality Electronic Design, ISQED 2012. 2012. p. 617-624. 6187557 https://doi.org/10.1109/ISQED.2012.6187557
Ravishankar, Chirag ; Ananthanarayanan, Sundaram ; Garg, Siddharth ; Kennings, Andrew. / Analysis and evaluation of greedy thread swapping based dynamic power management for MPSoC platforms. Proceedings of the 13th International Symposium on Quality Electronic Design, ISQED 2012. 2012. pp. 617-624
@inproceedings{f42518ac3fd0461d85014f38c81cc3a8,
title = "Analysis and evaluation of greedy thread swapping based dynamic power management for MPSoC platforms",
abstract = "Thread migration (TM) is a recently proposed dynamic power management technique for heterogeneous multi-processor system-on-chip (MPSoC) platforms that eliminates the area and power overheads incurred by fine-grained dynamic voltage and frequency scaling (DVFS) based power management. In this paper, we take the first step towards formally analyzing and experimentally evaluating the use of power-aware TM for parallel data streaming applications on MPSoC platforms. From an analysis perspective, we characterize the optimal mapping of threads to cores and prove the convergence properties of a complexity effective greedy thread swapping based TM algorithm to the globally optimal solution. The proposed techniques are evaluated on a 9-core FPGA based MPSoC prototype equipped with fully-functional TM and DVFS support, and running a parallelized video encoding benchmark based on the Motion Picture Experts Group (MPEG-2) standard. Our experimental results validate the proposed theoretical analysis, and show that the proposed TM algorithm provides within 8{\%} of the DVFS performance under the same power budget, and assuming no overheads for DVFS. Assuming voltage regulator inefficiency of 80{\%}, the proposed TM algorithm has 9{\%} higher performance than DVFS, again under the same total power budget.",
keywords = "DVFS, FPGA, Multi-core, Power management, Thread migration",
author = "Chirag Ravishankar and Sundaram Ananthanarayanan and Siddharth Garg and Andrew Kennings",
year = "2012",
doi = "10.1109/ISQED.2012.6187557",
language = "English (US)",
isbn = "9781467310369",
pages = "617--624",
booktitle = "Proceedings of the 13th International Symposium on Quality Electronic Design, ISQED 2012",

}

TY - GEN

T1 - Analysis and evaluation of greedy thread swapping based dynamic power management for MPSoC platforms

AU - Ravishankar, Chirag

AU - Ananthanarayanan, Sundaram

AU - Garg, Siddharth

AU - Kennings, Andrew

PY - 2012

Y1 - 2012

N2 - Thread migration (TM) is a recently proposed dynamic power management technique for heterogeneous multi-processor system-on-chip (MPSoC) platforms that eliminates the area and power overheads incurred by fine-grained dynamic voltage and frequency scaling (DVFS) based power management. In this paper, we take the first step towards formally analyzing and experimentally evaluating the use of power-aware TM for parallel data streaming applications on MPSoC platforms. From an analysis perspective, we characterize the optimal mapping of threads to cores and prove the convergence properties of a complexity effective greedy thread swapping based TM algorithm to the globally optimal solution. The proposed techniques are evaluated on a 9-core FPGA based MPSoC prototype equipped with fully-functional TM and DVFS support, and running a parallelized video encoding benchmark based on the Motion Picture Experts Group (MPEG-2) standard. Our experimental results validate the proposed theoretical analysis, and show that the proposed TM algorithm provides within 8% of the DVFS performance under the same power budget, and assuming no overheads for DVFS. Assuming voltage regulator inefficiency of 80%, the proposed TM algorithm has 9% higher performance than DVFS, again under the same total power budget.

AB - Thread migration (TM) is a recently proposed dynamic power management technique for heterogeneous multi-processor system-on-chip (MPSoC) platforms that eliminates the area and power overheads incurred by fine-grained dynamic voltage and frequency scaling (DVFS) based power management. In this paper, we take the first step towards formally analyzing and experimentally evaluating the use of power-aware TM for parallel data streaming applications on MPSoC platforms. From an analysis perspective, we characterize the optimal mapping of threads to cores and prove the convergence properties of a complexity effective greedy thread swapping based TM algorithm to the globally optimal solution. The proposed techniques are evaluated on a 9-core FPGA based MPSoC prototype equipped with fully-functional TM and DVFS support, and running a parallelized video encoding benchmark based on the Motion Picture Experts Group (MPEG-2) standard. Our experimental results validate the proposed theoretical analysis, and show that the proposed TM algorithm provides within 8% of the DVFS performance under the same power budget, and assuming no overheads for DVFS. Assuming voltage regulator inefficiency of 80%, the proposed TM algorithm has 9% higher performance than DVFS, again under the same total power budget.

KW - DVFS

KW - FPGA

KW - Multi-core

KW - Power management

KW - Thread migration

UR - http://www.scopus.com/inward/record.url?scp=84863643182&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863643182&partnerID=8YFLogxK

U2 - 10.1109/ISQED.2012.6187557

DO - 10.1109/ISQED.2012.6187557

M3 - Conference contribution

SN - 9781467310369

SP - 617

EP - 624

BT - Proceedings of the 13th International Symposium on Quality Electronic Design, ISQED 2012

ER -