Half-DRAM: a High-bandwidth and Low-power DRAM Architecture from the Rethinking of Fine-grained Activation

被引:0
|
作者
Zhang, Tao [1 ,2 ]
Chen, Ke [3 ]
Xu, Cong [1 ]
Sun, Guangyu [4 ]
Wang, Tao [4 ]
Xie, Yuan [1 ]
机构
[1] Penn State Univ, University Pk, PA 16802 USA
[2] NVIDIA Corp, Santa Clara, CA USA
[3] Oracle Corp, Santa Clara, CA USA
[4] Peking Univ, Beijing, Peoples R China
基金
美国国家科学基金会; 中国国家自然科学基金;
关键词
MEMORY; ENERGY; REFRESH; DESIGN;
D O I
暂无
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
DRAM memory is a major contributor for the total power consumption in modern computing systems. Consequently, power reduction for DRAM memory is critical to improve system-level power efficiency. Fine-grained DRAM architecture [1, 2] has been proposed to reduce the activation/precharge power However those prior work either incurs significant performance degradation or introduces large area overhead. In this paper; we propose a novel memory architecture Half-DRAM, in which the DRAM array is reorganized to enable only half of a row being activated. The half-row activation can effectively reduce activation power and meanwhile sustain the full bandwidth one bank can provide. In addition, the half: row activation in Half-DRAM relaxes the power constraint in DRAM, and opens up opportunities for further performance gain. Furthermore, two half: row accesses can be issued in parallel by integrating the sub-array level parallelism to improve the memory level parallelism. The experimental results show that Half-DRAM can achieve both significant performance improvement and power reduction, with negligible design overhead.
引用
收藏
页码:349 / 360
页数:12
相关论文
共 50 条
  • [1] FINE-GRAINED ACTIVATION FOR POWER REDUCTION IN DRAM
    Cooper-Balis, Elliott
    Jacob, Bruce
    IEEE MICRO, 2010, 30 (03) : 34 - 47
  • [2] Fine-Grained DRAM: Energy-Efficient DRAM for Extreme Bandwidth Systems
    O'Connor, Mike
    Chatterjee, Niladrish
    Lee, Donghyuk
    Wilson, John
    Agrawal, Aditya
    Keckler, Stephen W.
    Dally, William J.
    50TH ANNUAL IEEE/ACM INTERNATIONAL SYMPOSIUM ON MICROARCHITECTURE (MICRO), 2017, : 41 - 54
  • [3] Sectored DRAM: A Practical Energy-Efficient and High-Performance Fine-Grained DRAM Architecture
    Olgun, Ataberk
    Bostanci, F. Nisa
    de Oliveira Junior, Geraldo Francisco
    Tugrul, Yahya Can
    Ul Bera, Rah
    Yaglikci, Abdullah Giray
    Hassan, Hasan
    Ergin, Oguz
    Mutlu, Onur
    ACM TRANSACTIONS ON ARCHITECTURE AND CODE OPTIMIZATION, 2024, 21 (03)
  • [4] Fine-Grained Data Management for DRAM/SSD Hybrid Main Memory Architecture
    Wang, Liyu
    Wang, Qiang
    Chen, Lan
    Hao, Xiaoran
    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2016, E99D (12): : 3172 - 3176
  • [5] Partial Row Activation for Low-Power DRAM System
    Lee, Yebin
    Kim, Hyeonggyu
    Hong, Seokin
    Kim, Soontae
    2017 23RD IEEE INTERNATIONAL SYMPOSIUM ON HIGH PERFORMANCE COMPUTER ARCHITECTURE (HPCA), 2017, : 217 - 228
  • [6] High-performance embedded SOI DRAM architecture for the low-power supply
    Yamauchi, T
    Morisita, F
    Maeda, S
    Arimoto, K
    Fujishima, K
    Ozaki, H
    Yoshihara, T
    IEEE JOURNAL OF SOLID-STATE CIRCUITS, 2000, 35 (08) : 1169 - 1178
  • [7] Low-Power and High-Speed DRAM Readout Scheme
    Sharroush, Sherif M.
    2013 IEEE 20TH INTERNATIONAL CONFERENCE ON ELECTRONICS, CIRCUITS, AND SYSTEMS (ICECS), 2013, : 791 - 794
  • [8] Controller Architecture for Low-Power, Low-Latency DRAM With Built-in Cache
    Liu, Zhi-Yong
    Shih, Hsiu-Chuan
    Lin, Bing-Yang
    Wu, Cheng-Wen
    IEEE DESIGN & TEST, 2017, 34 (02) : 69 - 78
  • [9] Photonic integrated chips for low-power, high-bandwidth communications
    Komljenovic, Tin
    Zhang, Chong
    Zhang, Shangjian
    Bowers, John E.
    2017 IEEE PHOTONICS SOCIETY SUMMER TOPICAL MEETING SERIES (SUM), 2017, : 135 - 136
  • [10] Heterogeneous Multi-Channel: Fine-Grained DRAM Control for Both System Performance and Power Efficiency
    Zhang, Guangfei
    Wang, Huandong
    Chen, Xinke
    Huang, Shuai
    Li, Peng
    2012 49TH ACM/EDAC/IEEE DESIGN AUTOMATION CONFERENCE (DAC), 2012, : 876 - 881