Behavior-Aware cache hierarchy optimization for low-power multi-core embedded systems

Huatao Zhao, Xiao Luo, Chen Zhu, Takahiro Watanabe, Tianbo Zhu

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In modern embedded systems, the increasing number of cores requires efficient cache hierarchies to ensure data throughput, but such cache hierarchies are restricted by their tumid size and interference accesses which leads to both performance degradation and wasted energy. In this paper, we firstly propose a behavior-Aware cache hierarchy (BACH) which can optimally allocate the multi-level cache resources to many cores and highly improved the efficiency of cache hierarchy, resulting in low energy consumption. The BACH takes full advantage of the explored application behaviors and runtime cache resource demands as the cache allocation bases, so that we can optimally configure the cache hierarchy to meet the runtime demand. The BACH was implemented on the GEM5 simulator. The experimental results show that energy consumption of a three-level cache hierarchy can be saved from 5.29% up to 27.94% compared with other key approaches while the performance of the multi-core system even has a slight improvement counting in hardware overhead.

Original languageEnglish
Article number1740067
JournalModern Physics Letters B
Volume31
Issue number19-21
DOIs
Publication statusPublished - 2017 Jul 30

Fingerprint

hierarchies
optimization
energy consumption
resources
simulators
counting
hardware
degradation
interference

Keywords

  • Cache hierarchy
  • embedded system
  • low power
  • multi-core processor

ASJC Scopus subject areas

  • Statistical and Nonlinear Physics
  • Condensed Matter Physics

Cite this

Behavior-Aware cache hierarchy optimization for low-power multi-core embedded systems. / Zhao, Huatao; Luo, Xiao; Zhu, Chen; Watanabe, Takahiro; Zhu, Tianbo.

In: Modern Physics Letters B, Vol. 31, No. 19-21, 1740067, 30.07.2017.

Research output: Contribution to journalArticle

Zhao, Huatao ; Luo, Xiao ; Zhu, Chen ; Watanabe, Takahiro ; Zhu, Tianbo. / Behavior-Aware cache hierarchy optimization for low-power multi-core embedded systems. In: Modern Physics Letters B. 2017 ; Vol. 31, No. 19-21.
@article{54ba02f8f7ab4022ac4b834c3571a0a3,
title = "Behavior-Aware cache hierarchy optimization for low-power multi-core embedded systems",
abstract = "In modern embedded systems, the increasing number of cores requires efficient cache hierarchies to ensure data throughput, but such cache hierarchies are restricted by their tumid size and interference accesses which leads to both performance degradation and wasted energy. In this paper, we firstly propose a behavior-Aware cache hierarchy (BACH) which can optimally allocate the multi-level cache resources to many cores and highly improved the efficiency of cache hierarchy, resulting in low energy consumption. The BACH takes full advantage of the explored application behaviors and runtime cache resource demands as the cache allocation bases, so that we can optimally configure the cache hierarchy to meet the runtime demand. The BACH was implemented on the GEM5 simulator. The experimental results show that energy consumption of a three-level cache hierarchy can be saved from 5.29{\%} up to 27.94{\%} compared with other key approaches while the performance of the multi-core system even has a slight improvement counting in hardware overhead.",
keywords = "Cache hierarchy, embedded system, low power, multi-core processor",
author = "Huatao Zhao and Xiao Luo and Chen Zhu and Takahiro Watanabe and Tianbo Zhu",
year = "2017",
month = "7",
day = "30",
doi = "10.1142/S021798491740067X",
language = "English",
volume = "31",
journal = "Modern Physics Letters B",
issn = "0217-9849",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "19-21",

}

TY - JOUR

T1 - Behavior-Aware cache hierarchy optimization for low-power multi-core embedded systems

AU - Zhao, Huatao

AU - Luo, Xiao

AU - Zhu, Chen

AU - Watanabe, Takahiro

AU - Zhu, Tianbo

PY - 2017/7/30

Y1 - 2017/7/30

N2 - In modern embedded systems, the increasing number of cores requires efficient cache hierarchies to ensure data throughput, but such cache hierarchies are restricted by their tumid size and interference accesses which leads to both performance degradation and wasted energy. In this paper, we firstly propose a behavior-Aware cache hierarchy (BACH) which can optimally allocate the multi-level cache resources to many cores and highly improved the efficiency of cache hierarchy, resulting in low energy consumption. The BACH takes full advantage of the explored application behaviors and runtime cache resource demands as the cache allocation bases, so that we can optimally configure the cache hierarchy to meet the runtime demand. The BACH was implemented on the GEM5 simulator. The experimental results show that energy consumption of a three-level cache hierarchy can be saved from 5.29% up to 27.94% compared with other key approaches while the performance of the multi-core system even has a slight improvement counting in hardware overhead.

AB - In modern embedded systems, the increasing number of cores requires efficient cache hierarchies to ensure data throughput, but such cache hierarchies are restricted by their tumid size and interference accesses which leads to both performance degradation and wasted energy. In this paper, we firstly propose a behavior-Aware cache hierarchy (BACH) which can optimally allocate the multi-level cache resources to many cores and highly improved the efficiency of cache hierarchy, resulting in low energy consumption. The BACH takes full advantage of the explored application behaviors and runtime cache resource demands as the cache allocation bases, so that we can optimally configure the cache hierarchy to meet the runtime demand. The BACH was implemented on the GEM5 simulator. The experimental results show that energy consumption of a three-level cache hierarchy can be saved from 5.29% up to 27.94% compared with other key approaches while the performance of the multi-core system even has a slight improvement counting in hardware overhead.

KW - Cache hierarchy

KW - embedded system

KW - low power

KW - multi-core processor

UR - http://www.scopus.com/inward/record.url?scp=85026324337&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85026324337&partnerID=8YFLogxK

U2 - 10.1142/S021798491740067X

DO - 10.1142/S021798491740067X

M3 - Article

VL - 31

JO - Modern Physics Letters B

JF - Modern Physics Letters B

SN - 0217-9849

IS - 19-21

M1 - 1740067

ER -