Thermal-aware, heterogeneous materials for improved energy and reliability in 3D PCM architectures

被引:0
|
作者
Saadeldeen, Heba [1 ]
Deng, Zhaoxia [1 ]
Sherwood, Timothy [1 ]
Chong, Frederic T. [2 ]
机构
[1] Univ Calif Santa Barbara, Santa Barbara, CA 93106 USA
[2] Univ Chicago, Chicago, IL 60637 USA
关键词
PCM; 3D stack memory architecture; Thermal-aware; phase change materials; PHASE-CHANGE MEMORY; RANDOM-ACCESS MEMORY; DIE-STACKING; DESIGN; HARD;
D O I
10.1145/3132402.3132407
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
The properties of Phase-Change Memory (PCM) are defined in large part by the different chalcogenide materials employed. As the GeTe and Sb2Te3 ratios in the materials are changed, the operating temperatures needed for the phase change are also variable. Motivated by this phenomenon, we study the potential of exploiting different material compositions to achieve different trade-offs among the optimal operating temperatures, energy efficiency, write endurance and write latency. Specifically, we study the trade-offs for energy efficiency and lifetime in the scenario of using PCM materials for all layers of a 3D stack memory. Rather than a "one-memory-fitsall" approach, we propose Heterogeneous 3D PCM architectures by tailoring the Ge-Sb-Te ratios of PCM in concert with both the location and the intended function of these memories within the 3D stack. By varying the material compositions and their operating temperatures in correspondent with the non-uniform heat distribution across the stack, the heterogeneous PCM architectures improve the programming energy by up to 3.5X compared to the best homogeneous configuration. Moreover, the diversity in material compositions can also be exploited to protect error-correcting codes (ECC) by storing them in PCM materials with lower operating temperatures, which drastically reduces ECC early failures and brings a 30% improvement in the lifetime of the entire memory system. This architectural study attempts to make the case for exploring the whole material spectrum and the manufacturing cost associated with that.
引用
收藏
页码:223 / 236
页数:14
相关论文
共 50 条
  • [1] Thermal-Aware Floorplanning for 3D MPSoCs
    Ayala, Jose L.
    IEEE DESIGN & TEST OF COMPUTERS, 2011, 28 (02): : 78 - 78
  • [2] Thermal-aware incremental floorplanning for 3D ICs
    Li, Xin
    Ma, Yuchun
    Hong, Xianlong
    Dong, Sheqin
    ASICON 2007: 2007 7TH INTERNATIONAL CONFERENCE ON ASIC, VOLS 1 AND 2, PROCEEDINGS, 2007, : 1092 - 1095
  • [3] Thermal-aware detour routing in 3D NoCs
    Mukherjee, Priyajit
    Chatterjee, Navonil
    Chattopadhyay, Santanu
    JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2020, 144 : 230 - 245
  • [4] Interconnect and thermal-aware floorplanning for 3D microprocessors
    Hung, W. -L.
    Link, G. M.
    Xie, Yuan
    Vijaykrishnan, N.
    Irwin, M. J.
    ISQED 2006: PROCEEDINGS OF THE 7TH INTERNATIONAL SYMPOSIUM ON QUALITY ELECTRONIC DESIGN, 2006, : 98 - +
  • [5] Thermal-Aware Memory Mapping in 3D Designs
    Hsieh, Ang-Chih
    Hwang, TingTing
    DATE: 2009 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION, VOLS 1-3, 2009, : 1361 - 1366
  • [6] Thermal-aware task mapping for communication energy minimization on 3D NoC
    Shen, Lili
    Wu, Ning
    Yan, Gaizhen
    Ge, Fen
    IEICE ELECTRONICS EXPRESS, 2017, 14 (22):
  • [7] Thermal-Aware Memory Mapping in 3D Designs
    Hsieh, Ang-Chih
    Hwang, Tingting
    ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS, 2013, 13 (01)
  • [8] Thermal-aware application scheduling on device-heterogeneous embedded architectures
    Swaminathan, Karthik
    Kotra, Jagadish
    Liu, Huichu
    Sampson, Jack
    Kandemir, Mahmut
    Narayanan, Vijaykrishnan
    2015 28TH INTERNATIONAL CONFERENCE ON VLSI DESIGN (VLSID), 2015, : 221 - 226
  • [9] 3D thermal-aware floorplanner using a MOEA approximation
    Cuesta, David
    Risco-Martin, Jose L.
    Ayala, Jose L.
    Hidalgo, J. Ignacio
    INTEGRATION-THE VLSI JOURNAL, 2013, 46 (01) : 10 - 21
  • [10] Thermal-Aware Task Scheduling for 3D Multicore Processors
    Zhou, Xiuyi
    Yang, Jun
    Xu, Yi
    Zhang, Youtao
    Zhao, Jianhua
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2010, 21 (01) : 60 - 71