Exploiting bit-level write patterns to reduce energy consumption in hybrid cache architecture

被引:0
|
作者
Choi, Juhee [1 ]
Park, Heemin [2 ]
机构
[1] Sangmyung Univ, Dept Smart Informat Commun Engn, 31 Sangmyungdae Gil, Cheonan 31066, South Korea
[2] Sangmyung Univ, Dept Software, 31 Sangmyungdae Gil, Cheonan 31066, South Korea
基金
新加坡国家研究基金会;
关键词
non-volatile memory; STT-RAM; energy saving techniques; hybrid cache architecture; PERFORMANCE;
D O I
10.1587/elex.18.20210327
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
A hybrid cache architecture (HCA) is introduced to alleviate the drawbacks of non-volatile memory (NVM) technologies. Although researchers have offered meaningful ways to conserve energy, little attention has been paid to focus on write counts that are non-uniformly spread over a cache line. We propose a novel HCA to reduce the NVM write counts by exploiting bit-level write patterns. The data array is refined to separately store bits in the cache line to the NVM region and the SRAM region. As a result, 20.1% of energy is saved over prior works.
引用
收藏
页数:6
相关论文
共 50 条
  • [31] Prediction Hybrid Cache: An Energy-Efficient STT-RAM Cache Architecture
    Ahn, Junwhan
    Yoo, Sungjoo
    Choi, Kiyoung
    IEEE TRANSACTIONS ON COMPUTERS, 2016, 65 (03) : 940 - 951
  • [32] SLAM: High Performance and Energy Efficient Hybrid Last Level Cache Architecture for Multicore Embedded Systems
    Bhosale, Swapnil
    Pasricha, Sudeep
    2019 IEEE INTERNATIONAL CONFERENCE ON EMBEDDED SOFTWARE AND SYSTEMS (ICESS), 2019,
  • [33] A hybrid integrated architecture for energy consumption prediction
    Mate, Alejandro
    Peral, Jesus
    Ferrandez, Antonio
    Gil, David
    Trujillo, Juan
    FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE, 2016, 63 : 131 - 147
  • [34] S-FLASH: A NAND Flash-Based Deep Neural Network Accelerator Exploiting Bit-Level Sparsity
    Kang, Myeonggu
    Kim, Hyeonuk
    Shin, Hyein
    Sim, Jaehyeong
    Kim, Kyeonghan
    Kim, Lee-Sup
    IEEE TRANSACTIONS ON COMPUTERS, 2022, 71 (06) : 1291 - 1304
  • [35] BLADE: Energy-efficient attention accelerator with fused kernel and bit-level redundancy elimination
    Lin, Zhiwei
    Qin, Yubin
    Wang, Jiachen
    Wang, Yang
    Wang, Huanyu
    Zheng, Zhe
    Cui, Wenpeng
    Wei, Shaojun
    Hu, Yang
    Yin, Shouyi
    ELECTRONICS LETTERS, 2025, 61 (01)
  • [36] Low-Power Hybrid Magnetic Cache Architecture Exploiting Narrow-Width Values
    Imani, Mohsen
    Rahimi, Abbas
    Kim, Yeseong
    Rosing, Tajana
    2016 5TH NON-VOLATILE MEMORY SYSTEMS AND APPLICATIONS SYMPOSIUM (NVMSA), 2016,
  • [37] On-Chip Cache Architecture Exploiting Hybrid Memory Structures for Near-Threshold Computing
    Xu, Hongjie
    Shiomi, Jun
    Ishihara, Tohru
    Onodera, Hidetoshi
    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES, 2019, E102A (12) : 1741 - 1750
  • [38] A hybrid cooling system to reduce building energy consumption
    Guan, Lisa
    Bell, John
    Sharp, Alan
    Ball, Michael
    Gilbert, Dale
    FIRST INTERNATIONAL CONFERENCE ON BUILDING ENERGY AND ENVIRONMENT, PROCEEDINGS VOLS 1-3, 2008, : 533 - 540
  • [39] PAM: An Efficient Power-Aware Multi-level Cache Policy to Reduce Energy Consumption of Software Defined Network
    Meng, Xiaodong
    Zheng, Long
    Li, Li
    Li, Jie
    2015 1ST INTERNATIONAL CONFERENCE ON INDUSTRIAL NETWORKS AND INTELLIGENT SYSTEMS (INISCOM), 2015, : 18 - 23
  • [40] Towards CIM-friendly and Energy-Efficient DNN Accelerator via Bit-level Sparsity
    Karimzadeh, Foroozan
    Raychowdhury, Arijit
    PROCEEDINGS OF THE 2022 IFIP/IEEE 30TH INTERNATIONAL CONFERENCE ON VERY LARGE SCALE INTEGRATION (VLSI-SOC), 2022,