簡易檢索 / 詳目顯示

研究生: 周韓
Hang - Chou
論文名稱: Heal Unleveling: A Healing Scheme for Extending SSD Lifetime with Minimized Page-Copy Overhead
Heal Unleveling: A Healing Scheme for Extending SSD Lifetime with Minimized Page-Copy Overhead
指導教授: 謝仁偉
Jen-Wei Hsieh
口試委員: 吳晉賢
Chin-Hsien Wu
陳雅淑
Ya-Shu Chen
張立平
Li-Pin Chang
張哲維
Che-Wei Chang
學位類別: 碩士
Master
系所名稱: 電資學院 - 資訊工程系
Department of Computer Science and Information Engineering
論文出版年: 2017
畢業學年度: 105
語文別: 英文
論文頁數: 71
中文關鍵詞: NANDflashSSDhealing
外文關鍵詞: NAND flash SSD healing
相關次數: 點閱:151下載:0
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • The limited erase cycles of a block leads to the lifetime issue of SSD. One of
    studies is the healing technology, which is achieved by heating the cell and restore
    its state. To prolong the lifetime at most, one should even the healing count across
    all blocks. However, similar to erase operation, healing incurs extra writes. In this
    paper, we provide a healing scheme from a different perspective. We unleveling
    the usage between blocks and postpone the migration until certain point. Saved
    the copies by avoiding migration for leveling the heal count in early stage, the
    overall copies are therefore reduced. Through evaluation, the lifetime extension
    is up to 18% due to the great migration reduction. Also, it reduces the average
    heal time by 20 % at best.


    The limited erase cycles of a block leads to the lifetime issue of SSD. One of
    studies is the healing technology, which is achieved by heating the cell and restore
    its state. To prolong the lifetime at most, one should even the healing count across
    all blocks. However, similar to erase operation, healing incurs extra writes. In this
    paper, we provide a healing scheme from a different perspective. We unleveling
    the usage between blocks and postpone the migration until certain point. Saved
    the copies by avoiding migration for leveling the heal count in early stage, the
    overall copies are therefore reduced. Through evaluation, the lifetime extension
    is up to 18% due to the great migration reduction. Also, it reduces the average
    heal time by 20 % at best.

    1 Introduction 6 2 Background and Related Works 9 2.1 NAND Flash Memory . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2 Self-Healing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.3 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3 Heal Unleveling Design 14 3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 3.2 Group Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3.2.1 Free Group . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3.2.2 Normal Group . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.2.3 Working Group . . . . . . . . . . . . . . . . . . . . . . . . 19 3.2.4 Retired Group . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.3 Heal-unleveling Scheme . . . . . . . . . . . . . . . . . . . . . . . . 21 3.3.1 Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.3.2 Allocation Flow . . . . . . . . . . . . . . . . . . . . . . . . 22 3.4 Garbage Collection . . . . . . . . . . . . . . . . . . . . . . . . . . 25 3.5 Dormant Migration . . . . . . . . . . . . . . . . . . . . . . . . . . 26 4 Performance Evaluation 29 4.1 Experiment Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 1 4.2 Experiment Results . . . . . . . . . . . . . . . . . . . . . . . . . . 30 4.2.1 Copies Reduction . . . . . . . . . . . . . . . . . . . . . . . 30 4.2.2 Write Requests . . . . . . . . . . . . . . . . . . . . . . . . 31 4.2.3 Healing Latency . . . . . . . . . . . . . . . . . . . . . . . . 31 4.2.4 Migrations Over Time . . . . . . . . . . . . . . . . . . . . 34 4.2.5 Healings Over Time . . . . . . . . . . . . . . . . . . . . . . 36 4.2.6 Heal Distribution . . . . . . . . . . . . . . . . . . . . . . . 36 4.2.7 Heal Cumulation . . . . . . . . . . . . . . . . . . . . . . . 62 5 Conclusion 64

    [1] L. P. Chang, “A hybrid approach to nand-flash-based solid-state disks,” in
    IEEE Transactions on Computers, Oct. 2010, pp. 1337–1349.
    [2] L. P. Chang and C. D. Du, “Design and implementation of an efficient wearleveling algorithm for solid-state-disk microcontrollers,” in ACM Transactions on Design Automation of Electronic Systems, vol. 15, no. 6, Dec. 2009.
    [3] UMassTraceRepository, “Oltp application i/o.” [Online]. Available:
    http://traces.cs.umass.edu/index.php/Storage/Storage
    [4] Y. T. Chiu, “Forever flash,” in IEEE Spectrum, vol. 49, 2012, pp. 11–12.
    [5] N. Mielke, H. P. Belgal, A. Fazio, Q. Meng, and N. Righos, “Recovery effects in the distributed cycling of flash memories,” in IEEE International
    Reliability Physics Symposium Proceedings, 2006, pp. 29–35.
    [6] Q. Wu, G. Dong, and T. Zhang, “Exploiting heat-accelerated flash memory
    wear-out recovery to enable self-healing ssds,” in USENIX conference on Hot
    topics in storage and file systems (Hot Storage’11), 2011, pp. 4–4.
    [7] H.-T. Lue, P.-Y. Du, C.-P. Chen, W.-C. Chen, C.-C. Hsieh, Y.-H. Hsiao, Y.-
    H. Shih, and C.-Y. Lu, “Radically extending the cycling endurance of flash
    memory (to >100m cycles) by using built-in thermal annealing to self-heal
    the stress-induced damage,” in Electron Devices Meeting (IEDM’12), Dec.
    2012, pp. 9.1.1–9.1.4.
    65
    [8] V. Mohan, T. Siddiqua, S. Gurumurthi, and M. R. Stan, “How i learned to
    stop worrying and love flash endurance,” in Proceedings of the 2nd USENIX
    conference on Hot topics in storage and file systems, 2010.
    [9] R. Chen, Y. Wang, and Z. Shao, “Dheating: Dispersed heating repair for
    self-healing nand flash memory,” in Hardware/Software Codesign and System
    Synthesis (CODES+ISSS’13), Oct. 2013, pp. 1–10.
    [10] Y.-M. Chang, Y.-H. Chang, J.-J. Chen, T.-W. Kuo, H.-P. Li, and H.-T.
    Lue, “On trading wear-leveling with heal-leveling,” in Design Automation
    Conference (DAC’14), Jun. 2014, pp. 1–6.
    [11] L. P. Chang, “On efficient wear leveling for large-scale flash-memory storage
    systems,” in Proceedings of the 2007 ACM symposium on Applied computing,
    2007, pp. 1126–1130.
    [12] Y. H. Chang, J. W. Hsieh, and T. W. Kuo, “Improving flash wear-leveling
    by proactively moving static data,” in IEEE Transactions on Computers,
    vol. 59, 2010, pp. 53–65.
    [13] C. Wang and W.-F. Wong, “Observational wear leveling: An efficient algorithm for flash memory management,” in Design Automation Conference
    (DAC’12), 2012, pp. 235–242.
    [14] J. W. Hsieh, L. P. Chang, and T. W. Kuo, “Efficient identification of hot
    data for flash memory storage systems,” in ACM Transactions on Storage,
    vol. 2, no. 1, Feb. 2006, pp. 22–40.
    [15] M. L. Chiao and D. W. Chang, “Rose: A novel flash translation layer for nand
    flash memory based on hybrid address translation,” in IEEE Transactions on
    Computers, vol. 60, no. 6, Apr. 2011, pp. 753–765.
    66
    [16] M. Chiang, P. Lee, and R. Chang, “Managing flash memory in personal
    communication devices,” in Proceedings of the 1997 International Symposium
    on Consumer Electronics (ISCE97), 1997, pp. 177–182.
    [17] A. Kawaguchi, S. Nishioka, and H. Motoda, “A flash-memory based file system,” in Proceedings of the USENIX 1995 Technical Conference, Jan. 1995,
    pp. 155–164.
    [18] “Intel md516 nand flash memory js29f16g08aamc1,” 2007.

    無法下載圖示 全文公開日期 2022/02/13 (校內網路)
    全文公開日期 2027/02/13 (校外網路)
    全文公開日期 2027/02/13 (國家圖書館:臺灣博碩士論文系統)
    QR CODE