Dynamic Energy Management of FPGA Accelerators in Embedded Systems

Mohammad Hosseinabady; Jose Nunez-Yanez

doi:10.1145/3182172

Dynamic Energy Management of FPGA Accelerators in Embedded Systems

Mohammad Hosseinabady
, Jose Nunez-Yanez

Research output: Contribution to journal › Article (Academic Journal) › peer-review

8 Citations (Scopus)

928 Downloads (Pure)

Abstract

In this paper, we investigate how to utilize an FPGA in an embedded system to save energy. For this purpose, we study the energy efficiency of a hybrid FPGA-CPU device that can switch task execution between hardware and software with focus on periodic tasks. We have applied the proposed techniques to a robot map creation algorithm as a case study which shows up to 38\% energy reduction compared to using only the FPGA implementation. Overall, experimental results show up to 48\% energy reduction by applying the proposed DEM at runtime on thirteen individual tasks.

Original language	English
Article number	63
Number of pages	25
Journal	ACM Transactions on Embedded Computing Systems
Volume	17
Issue number	3
Early online date	1 Jun 2018
DOIs	https://doi.org/10.1145/3182172
Publication status	E-pub ahead of print - 1 Jun 2018

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

FPGA
Dynamic Voltage and Frequency Scaling
Dynamic Power Management
Zynq-SoC
Dynamic Energy Management

Access to Document

10.1145/3182172

Full-text PDF (accepted author manuscript)
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via ACM at https://dl.acm.org/citation.cfm?doid=3185335.3182172 . Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 4.38 MB

Handle.net

Persistent link

ENergy Efficient Adaptive Computing with multi-grain heterogeneous architectures (ENEAC)
Nunez-Yanez, J. L. (Principal Investigator)
5/01/16 → 4/01/20
Project: Research
Energy Proportional Computing With Heterogeneous and Reconfigurable Processors (ENPOWER)
Nunez-Yanez, J. L. (Principal Investigator)
1/11/13 → 30/04/17
Project: Research

Cite this

@article{63a6250c155c43739f103b240cf3c45e,

title = "Dynamic Energy Management of FPGA Accelerators in Embedded Systems",

abstract = "In this paper, we investigate how to utilize an FPGA in an embedded system to save energy. For this purpose, we study the energy efficiency of a hybrid FPGA-CPU device that can switch task execution between hardware and software with focus on periodic tasks. We have applied the proposed techniques to a robot map creation algorithm as a case study which shows up to 38\textbackslash{}\% energy reduction compared to using only the FPGA implementation. Overall, experimental results show up to 48\textbackslash{}\% energy reduction by applying the proposed DEM at runtime on thirteen individual tasks.",

keywords = "FPGA, Dynamic Voltage and Frequency Scaling, Dynamic Power Management, Zynq-SoC, Dynamic Energy Management",

author = "Mohammad Hosseinabady and Jose Nunez-Yanez",

year = "2018",

month = jun,

day = "1",

doi = "10.1145/3182172",

language = "English",

volume = "17",

journal = "ACM Transactions on Embedded Computing Systems",

issn = "1539-9087",

publisher = "Association for Computing Machinery",

number = "3",

}

TY - JOUR

T1 - Dynamic Energy Management of FPGA Accelerators in Embedded Systems

AU - Hosseinabady, Mohammad

AU - Nunez-Yanez, Jose

PY - 2018/6/1

Y1 - 2018/6/1

N2 - In this paper, we investigate how to utilize an FPGA in an embedded system to save energy. For this purpose, we study the energy efficiency of a hybrid FPGA-CPU device that can switch task execution between hardware and software with focus on periodic tasks. We have applied the proposed techniques to a robot map creation algorithm as a case study which shows up to 38\% energy reduction compared to using only the FPGA implementation. Overall, experimental results show up to 48\% energy reduction by applying the proposed DEM at runtime on thirteen individual tasks.

AB - In this paper, we investigate how to utilize an FPGA in an embedded system to save energy. For this purpose, we study the energy efficiency of a hybrid FPGA-CPU device that can switch task execution between hardware and software with focus on periodic tasks. We have applied the proposed techniques to a robot map creation algorithm as a case study which shows up to 38\% energy reduction compared to using only the FPGA implementation. Overall, experimental results show up to 48\% energy reduction by applying the proposed DEM at runtime on thirteen individual tasks.

KW - FPGA

KW - Dynamic Voltage and Frequency Scaling

KW - Dynamic Power Management

KW - Zynq-SoC

KW - Dynamic Energy Management

U2 - 10.1145/3182172

DO - 10.1145/3182172

M3 - Article (Academic Journal)

SN - 1539-9087

VL - 17

JO - ACM Transactions on Embedded Computing Systems

JF - ACM Transactions on Embedded Computing Systems

IS - 3

M1 - 63

ER -

Dynamic Energy Management of FPGA Accelerators in Embedded Systems

Abstract

UN SDGs

Keywords

Access to Document

Handle.net

Fingerprint

Projects

ENergy Efficient Adaptive Computing with multi-grain heterogeneous architectures (ENEAC)

Energy Proportional Computing With Heterogeneous and Reconfigurable Processors (ENPOWER)

Cite this