Locality-Driven Dynamic Flash Cache Allocation

被引:1
|
作者
Xu, Liang [1 ]
Xia, Qianbin [1 ]
Xiao, Weijun [1 ]
机构
[1] Virginia Commonwealth Univ, Dept Elect & Comp Engn, Richmond, VA 23284 USA
基金
美国国家科学基金会;
关键词
PERFORMANCE; STORAGE;
D O I
10.1109/DASC-PICom-DataCom-CyberSciTec.2017.43
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Flash-based SSDs are widely deployed as the storage caches to boost the system performance. However, unlike the traditional cache devices such as SRAM and DRAM, SSDs have internal garbage collection activities, which can severely degrade the cache performance. Moreover, SSD can only sustain limited P/E cycles. Therefore, traditional cache hit ratio oriented optimizations might not obtain the optimal performance for the SSD cache and can even shorten the device lifetime, especially for the SSD write cache, which could introduce more internal garbage collection processes and lifetime concerns. In this paper, we propose a reuse distance aware cache management to improve both the performance and lifetime of SSD-based write cache by compromising the cache hit ratio and the internal garbage collection overhead.
引用
收藏
页码:185 / 193
页数:9
相关论文
共 50 条
  • [1] Locality-driven MRC Construction and Cache Allocation
    Fu, Jianyu
    Arteaga, Dulcardo
    Zhao, Ming
    [J]. HPDC '18: PROCEEDINGS OF THE 27TH INTERNATIONAL SYMPOSIUM ON HIGH-PERFORMANCE PARALLEL AND DISTRIBUTED COMPUTING: POSTERS/DOCTORAL CONSORTIUM, 2018, : 19 - 20
  • [2] Locality-Driven Dynamic GPU Cache Bypassing
    Li, Chao
    Song, Shuaiwen Leon
    Dai, Hongwen
    Sidelnik, Albert
    Hari, Siva Kumar Sastry
    Zhou, Huiyang
    [J]. PROCEEDINGS OF THE 29TH ACM INTERNATIONAL CONFERENCE ON SUPERCOMPUTING (ICS'15), 2015, : 67 - 77
  • [3] Locality-Driven Architectural Cache Sub-banking for Leakage Energy Reduction
    Golubeva, Olga
    Loghi, Mirko
    Macii, Enrico
    Poncino, Massimo
    [J]. ISLPED'07: PROCEEDINGS OF THE 2007 INTERNATIONAL SYMPOSIUM ON LOW POWER ELECTRONICS AND DESIGN, 2007, : 274 - 279
  • [4] Locality Protected Dynamic Cache Allocation Scheme on GPUs
    Zhang, Yang
    Xing, Zuocheng
    Zhou, Li
    Zhu, Chunsheng
    [J]. 2016 IEEE TRUSTCOM/BIGDATASE/ISPA, 2016, : 1524 - 1530
  • [5] A locality-driven atomic group checkpoint protocol
    Wei, Zunce
    Li, Hon F.
    Goswami, Dhrubajyoti
    [J]. SEVENTH INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED COMPUTING, APPLICATIONS AND TECHNOLOGIES, PROCEEDINGS, 2006, : 558 - +
  • [6] Locality-Driven Parallel Power Grid Optimization
    Zeng, Zhiyu
    Li, Peng
    [J]. IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, 2009, 28 (08) : 1190 - 1200
  • [7] Locality-Driven Parallel Static Analysis for Power Delivery Networks
    Zeng, Zhiyu
    Feng, Zhuo
    Li, Peng
    Sarin, Vivek
    [J]. ACM TRANSACTIONS ON DESIGN AUTOMATION OF ELECTRONIC SYSTEMS, 2011, 16 (03)
  • [8] IMPROVING THE CACHE LOCALITY OF MEMORY ALLOCATION
    GRUNWALD, D
    ZORN, B
    HENDERSON, R
    [J]. SIGPLAN NOTICES, 1993, 28 (06): : 177 - 186
  • [9] Position Paper: Locality-Driven Scheduling of Tasks for Data-Dependent Multithreading
    Arteaga, Jaime
    Zuckerman, Stephane
    Garcia, Elkin
    Gao, Guang
    [J]. PROCEEDINGS OF 2014 IEEE INTERNATIONAL PARALLEL & DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW), 2014, : 1364 - 1368
  • [10] Locality-driven High-level I/O Aggregation for Processing Scientific Datasets
    Liu, Jialin
    Crysler, Bradly
    Lu, Yin
    Chen, Yong
    [J]. 2013 IEEE INTERNATIONAL CONFERENCE ON BIG DATA, 2013,