计算机科学技术学报 ›› 2019,Vol. 34 ›› Issue (1): 61-76.doi: 10.1007/s11390-019-1899-7

所属专题： Computer Architecture and Systems

• • 上一篇    下一篇

ROCO：用固态盘做缓存来改进主机端感知的叠瓦式磁记录磁盘的性能

Wen-Guo Liu1, Ling-Fang Zeng1,*, Senior Member, CCF, Member, ACM, IEEE Dan Feng1, Senior Member, CCF, Member, ACM, IEEE, and Kenneth B. Kent2, Senior Member, IEEE   

  1. 1 School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China;
    2 Faculty of Computer Science, University of New Brunswick, Fredericton, E3B 5A3, Canada
  • 收稿日期:2018-01-08 修回日期:2018-10-09 出版日期:2019-01-05 发布日期:2019-01-12
  • 通讯作者: Ling-Fang Zeng E-mail:lfzeng@hust.edu.cn
  • 作者简介:Wen-Guo Liu received his B.E. degree in information and computing science from Jinan University, Jinan, in 2008, and M.E. degree in computer science and technology from Hunan University, Changsha, in 2011. He is currently working toward his Ph.D. degree in the School of Computer Science and Technology from Huazhong University of Science and Technology (HUST), Wuhan. His research interest includes solid state drives, shingle magnetic recording, and hybrid storage systems.
  • 基金资助:
    This work was supported by the National Natural Science Foundation of China under Grant No. 61472153.

ROCO: Using a Solid State Drive Cache to Improve the Performance of a Host-Aware Shingled Magnetic Recording Drive

Wen-Guo Liu1, Ling-Fang Zeng1,*, Senior Member, CCF, Member, ACM, IEEE Dan Feng1, Senior Member, CCF, Member, ACM, IEEE, and Kenneth B. Kent2, Senior Member, IEEE   

  1. 1 School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China;
    2 Faculty of Computer Science, University of New Brunswick, Fredericton, E3B 5A3, Canada
  • Received:2018-01-08 Revised:2018-10-09 Online:2019-01-05 Published:2019-01-12
  • Contact: Ling-Fang Zeng E-mail:lfzeng@hust.edu.cn
  • About author:Wen-Guo Liu received his B.E. degree in information and computing science from Jinan University, Jinan, in 2008, and M.E. degree in computer science and technology from Hunan University, Changsha, in 2011. He is currently working toward his Ph.D. degree in the School of Computer Science and Technology from Huazhong University of Science and Technology (HUST), Wuhan. His research interest includes solid state drives, shingle magnetic recording, and hybrid storage systems.
  • Supported by:
    This work was supported by the National Natural Science Foundation of China under Grant No. 61472153.

PDF (PC)

442

补充材料

1

叠瓦式磁记录技术能够有效地提高硬盘的容量，由于具备较好的兼容性和新的针对叠瓦式磁记录技术的接口，主机端感知的叠瓦式磁记录技术会比其他类型的叠瓦式磁记录技术更受欢迎。但是，由于非顺序的写数据保存在磁盘缓存中，主机端感知的叠瓦式磁记录磁盘在写操作密集的工作负载下经常有性能下降的现象。非顺序的写操作主要有三个来源：更新写操作，小的随机写操作以及乱序写操作。本文提出了一种称为ROCO的混合存储系统，使用固态来做缓存来提高主机端感知的叠瓦式磁记录磁盘的性能。ROCO对属于同一数据区的乱序写操作重新排序，并使用固态盘来吸收更新数据和小的随机写数据。本文为固态盘缓存设计了一种称为CREA的缓存替换算法。CREA首先执行面向数据区的冷热数据识别，以此识别出缓存中冷的数据区和热的数据区，然后优先将能够顺序写或者通过主机端的读-更改写操作进行写的属于同一冷数据区的数据迁移到主机端感知的叠瓦式磁记录磁盘中。对于缓存中那些不能够顺序写的属于最热的数据区的数据，给以最低的数据替换优先级。实验结果表明，ROCO可以有效地减少写向主机端感知的叠瓦式磁记录磁盘的数据，并提高叠瓦式磁记录磁盘的性能。

关键词: 固态盘缓存, 主机端感知的叠瓦式磁记录磁盘, 面向数据区的数据块重新排序, 面向数据区的冷热数据识别, 数据替换算法

Abstract: Shingled magnetic recording (SMR) can effectively increase the capacity of hard disk drives (HDDs). Hostaware SMR (HA-SMR) is expected to be more popular than other SMR models because of its backward compatibility and new SMR-specific APIs. However, an HA-SMR drive often suffers performance degradation under write-intensive workloads because of frequent non-sequential writes buffered in the disk cache. The non-sequential writes mainly come from update writes, small random writes and out-of-order writes. In this paper, we propose a hybrid storage system called ROCO which aims to use a solid state drive (SSD) cache to improve the performance of an HA-SMR drive. ROCO reorders out-of-order writes belonging to the same zone and uses the SSD cache to absorb update writes and small random writes. We also design a data replacement algorithm called CREA for the SSD cache. CREA first conducts zone-oriented hot/cold data identification to identify cold-cached zones and hot-cached zones, and then evicts data blocks belonging to colder zones with higher priorities that can be sequentially written or written through host-side read-modify-write operations. It gives the lowest priority to data blocks belonging to the hottest-cached zone that have to be non-sequentially written. Experimental results show that ROCO can effectively reduce non-sequential writes to the HA-SMR drive and improve the performance of the HA-SMR drive.

Key words: solid state drive (SSD) cache, host-aware shingled magnetic recording (HA-SMR) drive, zone-oriented block reordering, zone-oriented hot/cold data identification, data replacement algorithm

[1] Wood R, Williams M, Kavcic A, Miles J. The feasibility of magnetic recording at 10 terabits per square inch on conventional media. IEEE Trans. Magnetics, 2009, 45(2):917-923.
[2] Venkataraman K S, Dong G, Zhang T. Techniques mitigating update-induced latency overhead in shingled magnetic recording. IEEE Trans. Magnetics, 2012, 48(5):1899-1905.
[3] Feldman T, Gibson G. Shingled magnetic recording areal density increase requires new data management. Login:the USENIX Magazine, 2013, 38(3):22-30.
[4] Zeng L F, Zhang Z H, Wang Y, Feng D, Kent K B. CosaFS:A cooperative shingle-aware file system. ACM Trans. Storage, 2017, 13(4):Article No. 34.
[5] Aghayev A, Shafaei M, Desnoyers P. Skylight-A window on shingled disk operation. ACM Trans. Storage, 2015, 11(4):Article No. 16.
[6] Wu F G, Yang M C, Fan Z Q, Zhang B Q, Ge X Z, Du D H. Evaluating host aware SMR drives. In Proc. the 8th USENIX Workshop on Hot Topics in Storage and File Systems, June 2016, pp.31-35.
[7] Wu F G, Fan Z Q, Yang M C, Zhang B Q, Ge X Z, Du D H. Performance evaluation of host aware shingled magnetic recording (HA-SMR) drives. IEEE Trans. Computers, 2017, 66(11):1932-1945.
[8] Wang C L, Wang D D, Chai Y P, Wang C W, Sun D S. Larger, cheaper, but faster:SSD-SMR hybrid storage boosted by a new SMR-oriented cache framework. In Proc. the 33rd Int. Conf. Massive Storage Systems and Technology, May 2017, pp.1-16.
[9] Chen F, Koufaty D A, Zhang X. Hystor:Making the best use of solid state drives in high performance storage systems. In Proc. the 25th Int. Conf. Supercomputing, May 2011, pp.22-32.
[10] Luo D, Wan J G, Zhu Y F, Zhao N N, Li F, Xie C S. Design and implementation of a hybrid shingled write disk system. IEEE Trans. Parallel and Distributed Systems, 2015, 27(4):1017-1029.
[11] Kim H, Shin D, Jeong Y et al. SHRD:Improving spatial locality in flash storage accesses by sequentializing in host and randomizing in device. In Proc. the 15th USENIX Conf. File and Storage Technologies, February 2017, pp.271-284.
[12] Bucy J S, Schindler J, Schlosser S W, Greger G R. The DiskSim simulation environment version 4.0 reference manual. Technical Report, Carnegie Mellon University, 2008. http://www.pdl.cmu.edu/DiskSim, May 2018.
[13] Narayanan D, Donnelly A, Rowstron A. Write off-loading:Practical power management for enterprise storage. ACM Trans. Storage, 2008, 4(3):Article No. 10.
[14] Cassuto Y, Sanvido M A A, Guyot C, Hall D R, Bandic Z Z. Indirection systems for shingled-recording disk drives. In Proc. the 26th IEEE Symp. Mass Storage Systems and Technologies, May 2010.
[15] Lin C I, Park D, He W P, Du D H. H-SWD:Incorporating hot data identification into shingled write disks. In Proc. the 20th IEEE Int. Symp. Modeling, Analysis and Simulation of Computer and Telecommunication Systems, August 2012, pp.321-330.
[16] Jones S N, Amer A, Miller E L, Long D D E, Pitchumani R, Strong C R. Classifying data to reduce long-term data movement in shingled write disks. ACM Trans. Storage, 2016, 12(1):Article No. 2.
[17] He W, Du D. SMaRT:An approach to shingled magnetic recording translation. In Proc. the 15th USENIX Conf. File and Storage Technologies, February 2017, pp.121-133.
[18] Xie T, Sun Y. Dynamic data reallocation in hybrid disk arrays. IEEE Trans. Parallel and Distributed Systems, 2010, 21(9):1330-1341.
[19] Mao B, Jiang H, Wu S Z, Tian L, Feng D, Chen J X, Zeng L F. HPDA:A hybrid parity-based disk array for enhanced performance and reliability. ACM Trans. Storage, 2012, 8(1):Article No. 4.
[20] Zeng L F, Feng D, Chen J X, Wei Q S, Veeravalli B, Liu W G. HRAID6ML:A hybrid RAID6 storage architecture with mirrored logging. In Proc. the 28th IEEE Symp. Mass Storage Systems and Technologies. April 2012.
[21] Jung H, Shim H, Park S, Kang S, Cha J. LRU-WSR:Integration of LRU and writes sequence reordering for flash memory. IEEE Trans. Consumer Electronics, 2008, 54(3):1215-1223.
[22] Park S, Jung D, Kang J, Kim J, Lee J. CFLRU:A replacement algorithm for flash memory. In Proc. the 2006 Int. Conf. Compilers, Architecture and Synthesis for Embedded Systems, October 2006, pp.234-241.
[23] Fan Z Q, Haghdoost A, Du D H, Voigt D. I/O-Cache:A non-volatile memory based buffer cache policy to improve storage performance. In Proc. the 23rd IEEE Int. Symp. Modeling, Analysis and Simulation of Computer and Telecommunication Systems, October 2015, pp.102-111.
[24] Fan Z, Wu F, Park D, Diehl J, Voigt D, Du D H. Hibachi:A cooperative hybrid cache with NVRAM and DRAM for storage arrays. In Proc. the 33rd IEEE Symp. Mass Storage Systems and Technologies, May 2017, pp.90-101.
[25] Park D, Fan Z Q, Nam Y J, Du D H. A lookahead read cache:Improving read performance for deduplication backup storage. Journal of Computer Science and Technology, 2017, 32(1):26-40.
No related articles found!
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] 刘明业; 洪恩宇;. Some Covering Problems and Their Solutions in Automatic Logic Synthesis Systems[J]. , 1986, 1(2): 83 -92 .
[2] 陈世华;. On the Structure of (Weak) Inverses of an (Weakly) Invertible Finite Automaton[J]. , 1986, 1(3): 92 -100 .
[3] 高庆狮; 张祥; 杨树范; 陈树清;. Vector Computer 757[J]. , 1986, 1(3): 1 -14 .
[4] 陈肇雄; 高庆狮;. A Substitution Based Model for the Implementation of PROLOG——The Design and Implementation of LPROLOG[J]. , 1986, 1(4): 17 -26 .
[5] 黄河燕;. A Parallel Implementation Model of HPARLOG[J]. , 1986, 1(4): 27 -38 .
[6] 闵应骅; 韩智德;. A Built-in Test Pattern Generator[J]. , 1986, 1(4): 62 -74 .
[7] 吴允曾;. On the Development of Applications of Logic in Programming[J]. , 1987, 2(1): 30 -34 .
[8] 唐同诰; 招兆铿;. Stack Method in Program Semantics[J]. , 1987, 2(1): 51 -63 .
[9] 衷仁保; 邢林; 任朝阳;. An Interactive System SDI on Microcomputer[J]. , 1987, 2(1): 64 -71 .
[10] 闵应骅;. Easy Test Generation PLAs[J]. , 1987, 2(1): 72 -80 .
版权所有 © 《计算机科学技术学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持：support@magtech.com.cn
总访问量：