Volatile STT-RAM scratchpad design and data allocation for low energy

Gabriel Rodríguez, Juan Tourinõ, Mahmut T. Kandemir

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

On-chip power consumption is one of the fundamental challenges of current technology scaling. Cache memories consume a sizable part of this power, particularly due to leakage energy. STT-RAM is one of several new memory technologies that have been proposed in order to improve power while preserving performance. It features high density and low leakage, but at the expense of write energy and performance. This article explores the use of STT-RAM-based scratchpad memories that trade nonvolatility in exchange for faster and less energetically expensive accesses, making them feasible for on-chip implementation in embedded systems. A novel multiretention scratchpad partitioning is proposed, featuring multiple storage spaces with different retention, energy, and performance characteristics. A customized compiler-based allocation algorithm suitable for use with such a scratchpad organization is described. Our experiments indicate that a multiretention STT-RAM scratchpad can provide energy savings of 53% with respect to an iso-area, hardware-managed SRAM cache.

Original languageEnglish (US)
Article number38
JournalACM Transactions on Architecture and Code Optimization
Volume11
Issue number4
DOIs
StatePublished - Dec 1 2014

Fingerprint

Random access storage
Data storage equipment
Cache memory
Static random access storage
Embedded systems
Energy conservation
Electric power utilization
Hardware
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems
  • Hardware and Architecture

Cite this

@article{9013f27f65f94773979f36226d58bc48,
title = "Volatile STT-RAM scratchpad design and data allocation for low energy",
abstract = "On-chip power consumption is one of the fundamental challenges of current technology scaling. Cache memories consume a sizable part of this power, particularly due to leakage energy. STT-RAM is one of several new memory technologies that have been proposed in order to improve power while preserving performance. It features high density and low leakage, but at the expense of write energy and performance. This article explores the use of STT-RAM-based scratchpad memories that trade nonvolatility in exchange for faster and less energetically expensive accesses, making them feasible for on-chip implementation in embedded systems. A novel multiretention scratchpad partitioning is proposed, featuring multiple storage spaces with different retention, energy, and performance characteristics. A customized compiler-based allocation algorithm suitable for use with such a scratchpad organization is described. Our experiments indicate that a multiretention STT-RAM scratchpad can provide energy savings of 53{\%} with respect to an iso-area, hardware-managed SRAM cache.",
author = "Gabriel Rodr{\'i}guez and Juan Tourin{\~o} and Kandemir, {Mahmut T.}",
year = "2014",
month = "12",
day = "1",
doi = "10.1145/2669556",
language = "English (US)",
volume = "11",
journal = "Transactions on Architecture and Code Optimization",
issn = "1544-3566",
publisher = "Association for Computing Machinery (ACM)",
number = "4",

}

Volatile STT-RAM scratchpad design and data allocation for low energy. / Rodríguez, Gabriel; Tourinõ, Juan; Kandemir, Mahmut T.

In: ACM Transactions on Architecture and Code Optimization, Vol. 11, No. 4, 38, 01.12.2014.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Volatile STT-RAM scratchpad design and data allocation for low energy

AU - Rodríguez, Gabriel

AU - Tourinõ, Juan

AU - Kandemir, Mahmut T.

PY - 2014/12/1

Y1 - 2014/12/1

N2 - On-chip power consumption is one of the fundamental challenges of current technology scaling. Cache memories consume a sizable part of this power, particularly due to leakage energy. STT-RAM is one of several new memory technologies that have been proposed in order to improve power while preserving performance. It features high density and low leakage, but at the expense of write energy and performance. This article explores the use of STT-RAM-based scratchpad memories that trade nonvolatility in exchange for faster and less energetically expensive accesses, making them feasible for on-chip implementation in embedded systems. A novel multiretention scratchpad partitioning is proposed, featuring multiple storage spaces with different retention, energy, and performance characteristics. A customized compiler-based allocation algorithm suitable for use with such a scratchpad organization is described. Our experiments indicate that a multiretention STT-RAM scratchpad can provide energy savings of 53% with respect to an iso-area, hardware-managed SRAM cache.

AB - On-chip power consumption is one of the fundamental challenges of current technology scaling. Cache memories consume a sizable part of this power, particularly due to leakage energy. STT-RAM is one of several new memory technologies that have been proposed in order to improve power while preserving performance. It features high density and low leakage, but at the expense of write energy and performance. This article explores the use of STT-RAM-based scratchpad memories that trade nonvolatility in exchange for faster and less energetically expensive accesses, making them feasible for on-chip implementation in embedded systems. A novel multiretention scratchpad partitioning is proposed, featuring multiple storage spaces with different retention, energy, and performance characteristics. A customized compiler-based allocation algorithm suitable for use with such a scratchpad organization is described. Our experiments indicate that a multiretention STT-RAM scratchpad can provide energy savings of 53% with respect to an iso-area, hardware-managed SRAM cache.

UR - http://www.scopus.com/inward/record.url?scp=84917685948&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84917685948&partnerID=8YFLogxK

U2 - 10.1145/2669556

DO - 10.1145/2669556

M3 - Article

AN - SCOPUS:84917685948

VL - 11

JO - Transactions on Architecture and Code Optimization

JF - Transactions on Architecture and Code Optimization

SN - 1544-3566

IS - 4

M1 - 38

ER -