Toward approximate computing for coarse-grained reconfigurable architectures

Omid Akbari; Mehdi Kamal; Ali Afzali-Kusha; Massoud Pedram; Muhammad Shafique

doi:10.1109/MM.2018.2873951

Toward approximate computing for coarse-grained reconfigurable architectures

Omid Akbari, Mehdi Kamal, Ali Afzali-Kusha, Massoud Pedram, Muhammad Shafique

Electrical Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

This paper presents a methodology for designing an approximate coarse- g rained reconfigurable architecture (X-CGRA), and its use for accelerating both error-resilient and error-sensitive applications. The output quality of the X-CGRA is manageable at the run-time for better performance and power/energy consumption tradeoffs. Results show up to 1.9× speedup and 2.3× lower energy consumption at the cost of 9.7% accuracy loss for the studied applications.

Original language	English (US)
Article number	8487001
Pages (from-to)	63-72
Number of pages	10
Journal	IEEE Micro
Volume	38
Issue number	6
DOIs	https://doi.org/10.1109/MM.2018.2873951
State	Published - Nov 1 2018

Keywords

Coarse-Grained Reconfigurable Architecture
Configurable Accuracy Computing
Energy-Efficiency

ASJC Scopus subject areas

Software
Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/MM.2018.2873951

Cite this

@article{0721ca85df7241afb61708ddaa44df0a,

title = "Toward approximate computing for coarse-grained reconfigurable architectures",

abstract = "This paper presents a methodology for designing an approximate coarse- g rained reconfigurable architecture (X-CGRA), and its use for accelerating both error-resilient and error-sensitive applications. The output quality of the X-CGRA is manageable at the run-time for better performance and power/energy consumption tradeoffs. Results show up to 1.9× speedup and 2.3× lower energy consumption at the cost of 9.7% accuracy loss for the studied applications.",

keywords = "Coarse-Grained Reconfigurable Architecture, Configurable Accuracy Computing, Energy-Efficiency",

author = "Omid Akbari and Mehdi Kamal and Ali Afzali-Kusha and Massoud Pedram and Muhammad Shafique",

note = "Funding Information: The work of O. Akbari, M. Kamal, and A. Afzali-Kusha was supported by the Iran National Science Foundation. Publisher Copyright: {\textcopyright} 1981-2012 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.",

year = "2018",

month = nov,

day = "1",

doi = "10.1109/MM.2018.2873951",

language = "English (US)",

volume = "38",

pages = "63--72",

journal = "IEEE Micro",

issn = "0272-1732",

publisher = "IEEE Computer Society",

number = "6",

}

TY - JOUR

T1 - Toward approximate computing for coarse-grained reconfigurable architectures

AU - Akbari, Omid

AU - Kamal, Mehdi

AU - Afzali-Kusha, Ali

AU - Pedram, Massoud

AU - Shafique, Muhammad

N1 - Funding Information: The work of O. Akbari, M. Kamal, and A. Afzali-Kusha was supported by the Iran National Science Foundation. Publisher Copyright: © 1981-2012 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.

PY - 2018/11/1

Y1 - 2018/11/1

N2 - This paper presents a methodology for designing an approximate coarse- g rained reconfigurable architecture (X-CGRA), and its use for accelerating both error-resilient and error-sensitive applications. The output quality of the X-CGRA is manageable at the run-time for better performance and power/energy consumption tradeoffs. Results show up to 1.9× speedup and 2.3× lower energy consumption at the cost of 9.7% accuracy loss for the studied applications.

AB - This paper presents a methodology for designing an approximate coarse- g rained reconfigurable architecture (X-CGRA), and its use for accelerating both error-resilient and error-sensitive applications. The output quality of the X-CGRA is manageable at the run-time for better performance and power/energy consumption tradeoffs. Results show up to 1.9× speedup and 2.3× lower energy consumption at the cost of 9.7% accuracy loss for the studied applications.

KW - Coarse-Grained Reconfigurable Architecture

KW - Configurable Accuracy Computing

KW - Energy-Efficiency

UR - http://www.scopus.com/inward/record.url?scp=85054642512&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054642512&partnerID=8YFLogxK

U2 - 10.1109/MM.2018.2873951

DO - 10.1109/MM.2018.2873951

M3 - Article

AN - SCOPUS:85054642512

SN - 0272-1732

VL - 38

SP - 63

EP - 72

JO - IEEE Micro

JF - IEEE Micro

IS - 6

M1 - 8487001

ER -

Toward approximate computing for coarse-grained reconfigurable architectures

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this