Droop Mitigating Last Level Cache Architecture for STTRAM

被引:0
作者
Aluru, Radha Krishna [1 ]
Ghosh, Swaroop [2 ]
机构
[1] Univ S Florida, Dept Comp Sci & Engn, Tampa, FL 33620 USA
[2] Penn State Univ, Sch Elect Engn & Comp Sci, University Pk, PA 16801 USA
来源
PROCEEDINGS OF THE 2017 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE) | 2017年
关键词
Droop; LLC; STTRAM; Bank; Latency; Energy; SPLASH benchmarks;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Spin-Transfer Torque Random Access Memory (STTRAM) is one of the emerging Non-Volatile Memory (NVM) technologies especially preferred for the Last Level Cache (LLC). The amount of current needed to switch the magnetization is high (similar to 100 mu A per bit). For a full cache line (512-bit) write, this extremely high current results in a voltage droop in the conventional cache architecture. Due to this droop, the write operation fails especially, when the farthest bank of the cache is accessed. In this paper, we propose a new cache architecture to mitigate this problem of droop and make the write operation successful. Instead of continuously writing the entire cache line (512-bit) in a single bank, the proposed architecture writes 64-bits in multiple physically separated locations across the cache. The simulation results obtained (both circuit and micro-architectural) comparing our proposed architecture against the conventional are found to be 1.96% (IPC) and 5.21% (energy).
引用
收藏
页码:262 / 265
页数:4
相关论文
共 11 条
  • [1] [Anonymous], NONV MEM WORKSH
  • [2] [Anonymous], 2009, TECH REP
  • [3] [Anonymous], 2013, ISPASS
  • [4] [Anonymous], 2015, HPCA
  • [5] [Anonymous], 2012, ISLPED
  • [6] [Anonymous], 2007, MICRO
  • [7] Chen E., 2010, IEEE T MAGNETICS
  • [8] Adaptive Write and Shift Current Modulation for Process Variation Tolerance in Domain Wall Caches
    Motaman, Seyedhamidreza
    Ghosh, Swaroop
    [J]. IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, 2016, 24 (03) : 944 - 953
  • [9] Tabrizi F, 2009, NONVOLATILE STT RAM
  • [10] Yarom Yuval., MAPPING INTEL LAST L