arrow
Return

Approximate Computing Using Multiple-Access Single-Charge Associative Memory

delete2018-07-01
delete33
delete
OA
AI
M
Mohsen Imani *
S
Shruti Patil
T
Tajana Rosing
DOI:10.1109/TETC.2016.2565262delete
deleteOriginal
deleteShare
deleteSave
View PDF
Abstract

Abstract

En 中文
Memory-based computing using associative memory is a promising way to reduce the energy consumption of important classes of streaming applications by avoiding redundant computations. A set of frequent patterns that represent basic functions are pre-stored in Ternary Content Addressable Memory (TCAM) and reused. The primary limitation to using associative memory in modern parallel processors is the large search energy required by TCAMs. In TCAMs, all rows that match, except hit rows, precharge and discharge for every search operation, resulting in high energy consumption. In this paper, we propose a new Multiple-Access Single-Charge (MASC) TCAM architecture which is capable of searching TCAM contents multiple times with only a single precharge cycle. In contrast to previous designs, the MASC TCAM keeps the matchline voltage of all miss-rows high and uses their charge for the next search operation, while only the hit rows discharge. We use periodic refresh to control the accuracy of the search. We also implement a new type of approximate associative memory by setting longer refresh times for MASC TCAMs, which yields search results within 1-2 bit Hamming distances of the exact value. To further decrease the energy consumption of MASC TCAM and reduce the area, we implement MASC with crossbar TCAMs. Our evaluation on AMD Southern Island GPU shows that using MASC (crossbar MASC) associative memory can improve the average floating point units energy efficiency by 33.4, 38.1, and 36.7 percent (37.7, 42.6, and 43.1 percent) for exact matching, selective 1-HD and 2-HD approximations respectively, providing an acceptable quality of service (PSNR > 30 dB and average relative error < 10 percent). This shows that MASC (crossbar MASC) can achieve 1.77X (1.93X) higher energy savings as compared to the state of the art implementation of GPGPU that uses voltage overscaling on TCAM.
Keywords:
Approximate computing
ternary content addressable memory
associative memory
non-volatile memory
GPUs

Journal

IEEE Transactions on Emerging Topics in Computing cover
IEEE Transactions on Emerging Topics in Computing
IF:
5.4
Papers:
1.1K
Citations:
4.0K

Organization

University of California System cover
University of California System
Scholars:
37.7W
Papers: 33.8W
Citations: 55.6W
Cited Papers

Cited Papers

Total Synthesis of Kendomycin Featuring Intramolecular Dötz Benzannulation
err2010-03-12
err0
PREAI
errKyosuke Tanaka; Masahito Watanabe; Kodai Ishibashi; Hiroshi Matsuyama; Yoko Saikawa; Masaya Nakata
errShare
errSave
Anthropometric Reference Data for Children and Adults: United States, 2003-2006
err2008-01-01
err0
errOAAI
errMargart A. McDowell; Cheryl D. Fryar; Cynthia L. Ogden; Katherine M. Flegal
errShare
errSave
Complementary resistive switches for passive nanocrossbar memories
err2010-04-18
err1.2K
PREAI
errLinn, Eike; Rosezin, Roland; Kuegeler, Carsten; Waser, Rainer
errShare
errSave
A Functional Hybrid Memristor Crossbar-Array/CMOS System for Data Storage and Neuromorphic Applications
err2011-12-09
err880
PREAI
errKim, Kuk-Hwan; Gaba, Siddharth; Wheeler, Dana; Cruz-Albrecht, Jose M.; Hussain, Tahir; Srinivasa, Narayan; Lu, Wei
errShare
errSave
Effects of a 12-Week Exercise Training Program on Physical Function in Institutionalized Frail Elderly
err2018-01-01
err0
errOAAI
errCristiane Batisti Ferreira; Pâmela dos Santos Teixeira; Geiane Alves dos Santos; Athila Teles Dantas Maya; Paula Americano do Brasil; Vinícius Carolino Souza; Cláudio Córdova; Aparecido Pimentel Ferreira; Ricardo Moreno Lima; Otávio de Toledo Nóbrega
errShare
errSave
Layered Learning
err2003-01-14
err0
errOAAI
errPeter Stone; Manuela Veloso
errShare
errSave
researcher View more