Towards Optimizing Refresh Energy in embedded-DRAM Caches using Private Blocks

被引:3
|
作者
Manohar, Sheel Sindhu [1 ]
Agarwal, Sukarn [1 ]
Kapoor, Hemangee K. [1 ]
机构
[1] Indian Inst Technol Guwahati, Gauhati 781039, Assam, India
关键词
embedded DRAM; Last Level Cache; Refresh Energy; Private Blocks; POWER;
D O I
10.1145/3299874.3317995
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
In recent years, the increased working set size of applications craves for more memory demand in terms of large size Last Level Caches (LLC). To fulfill this, embedded DRAM (eDRAM) caches have been considered as one of the best alternatives over conventional SRAM caches. eDRAM has a property of low leakage and provides more capacity in the same area footprint of SRAM. However, its retention period consumes significant refresh energy in the periodic refresh. In this paper, we present an approach to minimize the total energy spent on refreshes by considering the presence of private blocks in the LLC. Our approach restricts refreshing of those blocks that are loaded exclusively from the main memory on an LLC miss. Experimental result using full system simulation show 55% reduction in the total number of refreshes compared to baseline policy; and 62% reduction in total power consumption over SRAM.
引用
收藏
页码:225 / 230
页数:6
相关论文
共 13 条
  • [1] Refresh Optimised embedded-DRAM Caches based on Zero Data Detection
    Manohar, Sheel Sindhu
    Kapoor, Hemangee K.
    SAC '19: PROCEEDINGS OF THE 34TH ACM/SIGAPP SYMPOSIUM ON APPLIED COMPUTING, 2019, : 635 - 642
  • [2] Dynamic reconfiguration of embedded-DRAM caches employing zero data detection based refresh optimisation
    Manohar, Sheel Sindhu
    Kapoor, Hemangee K.
    JOURNAL OF SYSTEMS ARCHITECTURE, 2019, 100
  • [3] RANA: Towards Efficient Neural Acceleration with Refresh-Optimized Embedded DRAM
    Tu, Fengbin
    Wu, Weiwei
    Yin, Shouyi
    Liu, Leibo
    Wei, Shaojun
    2018 ACM/IEEE 45TH ANNUAL INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE (ISCA), 2018, : 340 - 352
  • [4] DRAM-based Coherent Caches and How to Take Advantage of the Coherence Protocol to Reduce the Refresh Energy
    Jaksic, Zoran
    Canal, Ramon
    2014 DESIGN, AUTOMATION AND TEST IN EUROPE CONFERENCE AND EXHIBITION (DATE), 2014,
  • [5] Low Power embedded DRAM Caches using BCH code Partitioning
    Reviriego, Pedro
    Sanchez-Macian, Alfonso
    Antonio Maestro, Juan
    2012 IEEE 18TH INTERNATIONAL ON-LINE TESTING SYMPOSIUM (IOLTS), 2012, : 79 - 83
  • [6] Low-Power High-Throughput LDPC Decoder Using Non-Refresh Embedded DRAM
    Park, Youn Sung
    Blaauw, David
    Sylvester, Dennis
    Zhang, Zhengya
    IEEE JOURNAL OF SOLID-STATE CIRCUITS, 2014, 49 (03) : 783 - 794
  • [7] Instruction fetch energy reduction using loop caches for embedded applications with small tight loops
    Motorola, Inc, Austin, TX, United States
    Proc Int Symp Low Power Electron Des Dig Tech Papers, (267-269):
  • [8] Energy Optimization in 3D MPSoCs with Wide-I/O DRAM Using Temperature Variation Aware Bank-wise Refresh
    Sadri, Mohammadsadegh
    Jung, Matthias
    Weis, Christian
    Wehn, Norbert
    Benini, Luca
    2014 DESIGN, AUTOMATION AND TEST IN EUROPE CONFERENCE AND EXHIBITION (DATE), 2014,
  • [9] Embedded DRAM (eDRAM) power-energy estimation using signal swing-based analytical model
    Park, YH
    Kook, J
    Yoo, HJ
    IEICE TRANSACTIONS ON ELECTRONICS, 2002, E85C (08): : 1664 - 1668
  • [10] Developing a spatial optimization design approach towards energy-saving and outdoor thermal comfortable densely-built residential blocks using a dynamic local energy balance model
    Liu, Lin
    Zhang, Zihong
    Lan, Shiying
    Tian, Xiaoyu
    Liu, Jing
    Liao, Wei
    Wang, Dan
    ENERGY AND BUILDINGS, 2025, 328