An optimization technique for low-energy embedded memory systems

Tadayuki Matsumura, Tohru Ishihara, Hiroto Yasuura

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

On-chip memories generally use higher supply (VDD) and higher threshold (Vth) voltages than those of logic parts to improve the static noise margin and to suppress the static energy consumption. However, the higher VDD increases the dynamic energy consumption. This paper proposes a hybrid memory architecture which consists of the following two regions; (1) a dynamic energy conscious region which uses low VDD and Vth and (2) a static energy conscious region which uses high V DD and Vth. The proposed architecture is applied to a scratchpad memory. This paper also proposes an optimization problem for finding the optimal code allocation and the memory configuration simultaneously, which minimizes the total energy consumption of the memory under constraints of a static noise margin (SNM), a write margin (WM) and a memory access delay. The memory configuration is defined by a memory division ratio, a β ratio and a VDD. Experimental results demonstrate that the total energy consumption of our original 90 nm SRAM can be reduced by 62.9% at the best case with a 4.56% area overhead without degradations of SNM, WM and access delay.

Original languageEnglish
Pages (from-to)239-249
Number of pages11
JournalIPSJ Transactions on System LSI Design Methodology
Volume2
DOIs
Publication statusPublished - Dec 1 2009

Fingerprint

Computer systems
Data storage equipment
Energy utilization
Memory architecture
Static random access storage
Threshold voltage
Degradation

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

An optimization technique for low-energy embedded memory systems. / Matsumura, Tadayuki; Ishihara, Tohru; Yasuura, Hiroto.

In: IPSJ Transactions on System LSI Design Methodology, Vol. 2, 01.12.2009, p. 239-249.

Research output: Contribution to journalArticle

@article{1ab3a0d3c97d48d585f0095c40690f90,
title = "An optimization technique for low-energy embedded memory systems",
abstract = "On-chip memories generally use higher supply (VDD) and higher threshold (Vth) voltages than those of logic parts to improve the static noise margin and to suppress the static energy consumption. However, the higher VDD increases the dynamic energy consumption. This paper proposes a hybrid memory architecture which consists of the following two regions; (1) a dynamic energy conscious region which uses low VDD and Vth and (2) a static energy conscious region which uses high V DD and Vth. The proposed architecture is applied to a scratchpad memory. This paper also proposes an optimization problem for finding the optimal code allocation and the memory configuration simultaneously, which minimizes the total energy consumption of the memory under constraints of a static noise margin (SNM), a write margin (WM) and a memory access delay. The memory configuration is defined by a memory division ratio, a β ratio and a VDD. Experimental results demonstrate that the total energy consumption of our original 90 nm SRAM can be reduced by 62.9{\%} at the best case with a 4.56{\%} area overhead without degradations of SNM, WM and access delay.",
author = "Tadayuki Matsumura and Tohru Ishihara and Hiroto Yasuura",
year = "2009",
month = "12",
day = "1",
doi = "10.2197/ipsjtsldm.2.239",
language = "English",
volume = "2",
pages = "239--249",
journal = "IPSJ Transactions on System LSI Design Methodology",
issn = "1882-6687",
publisher = "Information Processing Society of Japan",

}

TY - JOUR

T1 - An optimization technique for low-energy embedded memory systems

AU - Matsumura, Tadayuki

AU - Ishihara, Tohru

AU - Yasuura, Hiroto

PY - 2009/12/1

Y1 - 2009/12/1

N2 - On-chip memories generally use higher supply (VDD) and higher threshold (Vth) voltages than those of logic parts to improve the static noise margin and to suppress the static energy consumption. However, the higher VDD increases the dynamic energy consumption. This paper proposes a hybrid memory architecture which consists of the following two regions; (1) a dynamic energy conscious region which uses low VDD and Vth and (2) a static energy conscious region which uses high V DD and Vth. The proposed architecture is applied to a scratchpad memory. This paper also proposes an optimization problem for finding the optimal code allocation and the memory configuration simultaneously, which minimizes the total energy consumption of the memory under constraints of a static noise margin (SNM), a write margin (WM) and a memory access delay. The memory configuration is defined by a memory division ratio, a β ratio and a VDD. Experimental results demonstrate that the total energy consumption of our original 90 nm SRAM can be reduced by 62.9% at the best case with a 4.56% area overhead without degradations of SNM, WM and access delay.

AB - On-chip memories generally use higher supply (VDD) and higher threshold (Vth) voltages than those of logic parts to improve the static noise margin and to suppress the static energy consumption. However, the higher VDD increases the dynamic energy consumption. This paper proposes a hybrid memory architecture which consists of the following two regions; (1) a dynamic energy conscious region which uses low VDD and Vth and (2) a static energy conscious region which uses high V DD and Vth. The proposed architecture is applied to a scratchpad memory. This paper also proposes an optimization problem for finding the optimal code allocation and the memory configuration simultaneously, which minimizes the total energy consumption of the memory under constraints of a static noise margin (SNM), a write margin (WM) and a memory access delay. The memory configuration is defined by a memory division ratio, a β ratio and a VDD. Experimental results demonstrate that the total energy consumption of our original 90 nm SRAM can be reduced by 62.9% at the best case with a 4.56% area overhead without degradations of SNM, WM and access delay.

UR - http://www.scopus.com/inward/record.url?scp=79954482587&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79954482587&partnerID=8YFLogxK

U2 - 10.2197/ipsjtsldm.2.239

DO - 10.2197/ipsjtsldm.2.239

M3 - Article

AN - SCOPUS:79954482587

VL - 2

SP - 239

EP - 249

JO - IPSJ Transactions on System LSI Design Methodology

JF - IPSJ Transactions on System LSI Design Methodology

SN - 1882-6687

ER -