簡易檢索 / 詳目顯示

研究生: 劉宇軒
Yu-Hsuan Liu
論文名稱: IMRAID5: A New Layout to Reduce Rewriting Chances for RAID 5 Array with Interlaced Magnetic Recording
IMRAID5: A New Layout to Reduce Rewriting Chances for RAID 5 Array with Interlaced Magnetic Recording
指導教授: 謝仁偉
Jen-Wei Hsieh
口試委員: 陳雅淑
Ya-Shu Chen
吳晉賢
Chin-Hsien Wu
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: 2021
畢業學年度: 109
語文別: 英文
論文頁數: 56
中文關鍵詞: 交錯式磁紀錄容錯式磁碟陣列
外文關鍵詞: Redundant Array of Independent Disks
相關次數: 點閱:127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

  • As the demand of mass storage system rises in recent years, the market is
    looking for the storage media that has higher areal density. To increase the
    areal density, new technologies have been proposed such as shingled mag-
    netic recording(SMR) and interlaced magnetic recording(IMR). To further
    increase the space of storage system, a technology named redundant array of
    independent disks 5 (RAID 5) is used to take the bene ts from multiple disks
    storage system. RAID 5 utilizes data striping to improve the access speed
    and utilizes parity data to protect the system from one disk failure. Several
    previous papers have studied the potential of appplying SMRs to RAID 5.
    However, it shows that the solution provides bad access performance due to
    the costly RMW operation in SMR. In this paper, we propose our method
    named IMRAID5 to combine IMRs with RAID 5. To our best knowledge,
    we are the rst one to propose the solution of applying RAID 5 to IMR.
    Comparing to SMR, IMR has two advantages. First, a big portion of tracks
    can be freely rewritten without resulting in RMW operations. Second, the
    RMW overhead is xed and much smaller. Based on these two advantages,
    we expect the performance of RAID 5 IMR would be better than RAID 5
    SMR. Besides, IMRAID5 adopts two new technologies named virtual frame
    and top cache to help reduce the chances of RMW operation. These methods
    further improve the access performance of RAID 5 IMR. According to our
    experiment results, comparing to the baseline, IMRAID5 can decrease the
    frequency of RMW operations as well as the write latency 102 to 103 times
    on average.

    1 Introduction 6 2 Background and Motivation 11 2.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.2 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.2.1 Basic IMR RAID 5 design . . . . . . . . . . . . . . . . 15 2.2.2 Challenges of IMR RAID 5 design . . . . . . . . . . . . 18 3 Design of IMRAID5 20 3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.2 Virtual Frame Manager . . . . . . . . . . . . . . . . . . . . . . 20 3.2.1 Virtual Frame . . . . . . . . . . . . . . . . . . . . . . . 20 3.2.2 Data Distribution In Virtual Frame . . . . . . . . . . . 22 3.2.3 The Correspondence Of The PBAs In Stripes . . . . . 26 3.2.4 Ready State And Unready State . . . . . . . . . . . . . 28 3.2.5 Data Flow . . . . . . . . . . . . . . . . . . . . . . . . . 30 3.3 Top Cache Manager . . . . . . . . . . . . . . . . . . . . . . . . 33 3.3.1 Top Cache . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.3.2 Aging Technique . . . . . . . . . . . . . . . . . . . . . 34 3.4 Mapping Design . . . . . . . . . . . . . . . . . . . . . . . . . . 36 3.5 Recovery Process . . . . . . . . . . . . . . . . . . . . . . . . . 41 4 Experiment 44 4.1 Experiment Setup . . . . . . . . . . . . . . . . . . . . . . . . . 44 4.2 Experiment Result . . . . . . . . . . . . . . . . . . . . . . . . 48 4.2.1 RMW counts . . . . . . . . . . . . . . . . . . . . . . . 48 4.2.2 Latency . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.2.3 Impacts of Disk Number on RMW Counts . . . . . . . 54 5 Conclusion 56

    [1] Q. M. Le, K. SathyanarayanaRaju, A. Amer, and J. Holliday, \Work-
    load impact on shingled write disks: All-writes can be alright," in 2011
    IEEE 19th Annual International Symposium on Modelling, Analysis,
    and Simulation of Computer and Telecommunication Systems, pp. 444{
    446, IEEE, 2011.
    [2] R. Wood, M. Williams, A. Kavcic, and J. Miles, \The feasibility of mag-
    netic recording at 10 terabits per square inch on conventional media,"
    IEEE Transactions on Magnetics, vol. 45, no. 2, pp. 917{923, 2009.
    [3] R. E. Rottmayer, S. Batra, D. Buechel, W. A. Challener, J. Hohlfeld,
    Y. Kubota, L. Li, B. Lu, C. Mihalcea, K. Mount eld, et al., \Heat-
    assisted magnetic recording," IEEE Transactions on Magnetics, vol. 42,
    no. 10, pp. 2417{2421, 2006.
    [4] M. H. Kryder, E. C. Gage, T. W. McDaniel, W. A. Challener, R. E.
    Rottmayer, G. Ju, Y.-T. Hsia, and M. F. Erden, \Heat assisted magnetic
    recording," Proceedings of the IEEE, vol. 96, no. 11, pp. 1810{1835,
    2008.
    [5] S. Granz, J. Jury, C. Rea, G. Ju, J. Thiele, T. Rausch, and E. C. Gage,
    \Areal density comparison between conventional, shingled, and inter-
    laced heat-assisted magnetic recording with multiple sensor magnetic
    recording," IEEE Transactions on Magnetics, vol. 55, no. 3, pp. 1{3,
    2019.
    [6] C. Rea, P. Krivosik, V. Venugopal, M. F. Erden, S. Stokes, P. Subedi,
    M. Cordle, M. Benakli, H. Zhou, D. Karns, D. Saunders, S. Franzen,G. Ju, T. Rausch, M. Seigler, and E. Gage, \Heat-assisted recording:
    Advances in recording integration," IEEE Transactions on Magnetics,
    vol. 53, no. 11, pp. 1{6, 2017.
    [7] F. Wu, B. Li, B. Zhang, Z. Cao, J. Diehl, H. Wen, and D. H. Du,
    \Tracklace: Data management for interlaced magnetic recording," IEEE
    Transactions on Computers, vol. 70, no. 3, pp. 347{358, 2021.
    [8] M. H. Hajkazemi, A. N. Kulkarni, P. Desnoyers, and T. R. Feldman,
    \Track-based translation layers for interlaced magnetic recording," in
    Proceedings of the 2019 USENIX Conference on Usenix Annual Techni-
    cal Conference, USENIX ATC '19, (USA), p. 821{832, USENIX Asso-
    ciation, 2019.
    [9] Y. Liang, M.-C. Yang, and S.-H. Chen, \Magic: Making imr based hdd
    perform like cmr based hdd," IEEE Transactions on Computers, pp. 1{
    1, 2021.
    [10] W. Liu, D. Feng, L. Zeng, and J. Chen, \Understanding the swd-based
    raid system," in 2014 International Conference on Cloud Computing
    and Big Data, pp. 175{181, 2014.
    [11] Q. M. Le, A. Amer, and J. Holliday, \Smr disks for mass storage
    systems," in 2015 IEEE 23rd International Symposium on Modeling,
    Analysis, and Simulation of Computer and Telecommunication Systems,
    pp. 228{231, 2015.
    [12] S. Bhargava, Heat-Assisted Magnetic Recording: Fundamental Limits to
    Inverse Electromagnetic Design. PhD thesis, EECS Department, Uni-
    versity of California, Berkeley, May 2015.
    [13] A. Aghayev and P. Desnoyers, \Skylight|a window on shingled disk
    operation," in 13th USENIX Conference on File and Storage Technolo-
    gies (FAST 15), (Santa Clara, CA), pp. 135{149, USENIX Association,
    Feb. 2015.
    [14] C. Ma, Z. Shen, J. Wang, Y. Wang, R. Chen, Y. Guan, and Z. Shao,
    \Tiler: An autonomous region-based scheme for smr storage," IEEE
    Transactions on Computers, vol. 70, no. 2, pp. 291{304, 2021.
    [15] F. Wu, B. Li, Z. Cao, B. Zhang, M.-H. Yang, H. Wen, and D. H. Du,
    \Zonealloy: Elastic data and space management for hybrid SMR drives,"
    in 11th USENIX Workshop on Hot Topics in Storage and File Systems
    (HotStorage 19), (Renton, WA), USENIX Association, July 2019.
    [16] S. Granz, W. Zhu, E. C. S. Seng, U. H. Kan, C. Rea, G. Ju, J. Thiele,
    T. Rausch, and E. C. Gage, \Heat-assisted interlaced magnetic record-
    ing," IEEE Transactions on Magnetics, vol. 54, no. 2, pp. 1{4, 2018.
    [17] E. Hwang, J. Park, R. Rauschmayer, and B. Wilson, \Interlaced mag-
    netic recording," IEEE Transactions on Magnetics, vol. 53, no. 4, pp. 1{
    7, 2016.
    [18] D. Luo, T. Yao, X. Qu, J. Wan, and C. Xie, \Dvs: Dynamic variable-
    width striping raid for shingled write disks," in 2016 IEEE International
    Conference on Networking, Architecture and Storage (NAS), pp. 1{10,
    IEEE, 2016.
    [19] Z.-W. Lu and G. Zhou, \Design and implementation of hybrid shin-
    gled recording raid system," in 2016 IEEE 14th Intl Conf on De-
    pendable, Autonomic and Secure Computing, 14th Intl Conf on Per-
    vasive Intelligence and Computing, 2nd Intl Conf on Big Data Intel-
    ligence and Computing and Cyber Science and Technology Congress
    (DASC/PiCom/DataCom/CyberSciTech), pp. 937{942, IEEE, 2016.
    [20] E. K. Lee, \Software and performance issues in the implementation
    of a raid prototype," in CALIFORNIA UNIV BERKELEY DEPT OF
    ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, DE-
    FENSE TECHNICAL INFORMATION CENTER, 1990.
    [21] MSR Cambridge, \Msr cambridge trace," 2021. [Online; accessed 2011].
    [22] Wikipedia contributors, \Hard disk drive performance characteristics,"
    2021. [Online; accessed 20-Jan-2021].

    無法下載圖示 全文公開日期 2026/09/13 (校內網路)
    全文公開日期 本全文未授權公開 (校外網路)
    全文公開日期 本全文未授權公開 (國家圖書館:臺灣博碩士論文系統)
    QR CODE