Please wait a minute...
浙江大学学报(工学版)
J4  2012, Vol. 46 Issue (2): 294-300    DOI: 10.3785/j.issn.1008-973X.2012.02.018
    
An flash-based hybrid storage model for database
XU Chang, SHOU Li-dan, CHEN Gang, HU Tian-lei
College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China
Download:   PDF(0KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

 A novel hybrid storage model was proposed for the relational databases ,according to the prevalence and the IO characteristic of the flash disks. In this model, the storage media of the database consists of both magnetic and flash disk. The data in the database are deployed on the suitable media according to their access patterns. The model uses a mapping table to translate the logical accesses to the correspondent flash address. To improve the total system performance, the page deployment of the system is tuned adaptively according to the physical parameter of the devices and the recent access patterns, which are collected via the sliding windows. The experiments on the simulated OLTP workloads demonstrate that the model significantly improves the IO performance. The model is also proved to be extremely compliant with the various workloads.

Published: 20 March 2012
CLC:  TP 393.08  
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Cite this article:

XU Chang, SHOU Li-dan, CHEN Gang, HU Tian-lei. An flash-based hybrid storage model for database. J4, 2012, 46(2): 294-300.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2012.02.018     OR     http://www.zjujournals.com/eng/Y2012/V46/I2/294


一种基于闪存的数据库复合存储模型

 针对闪存硬盘的快速发展趋势以及存取特性,提出一种数据库复合存储模型.在该模型中,数据库同时包括闪存硬盘和磁性硬盘2种设备,数据根据访问特征被存储到不同的设备中.模型使用地址映射表来转换对闪存访问,并使用滑动窗口技术,通过最近一段时间内访问的统计信息和设备本身的物理特性,自适应地动态调整页面的分布,使数据库获得更高的性能.通过基于在线事务处理应用的性能仿真测试,结果表明，该复合模型可以显著提高I/O性能,并对工作集的变化有良好的适应性.

［1］ LAI S K. Flash memories: Successes and challenges ［J］. IBM Journal of Research and Development, 2008, 4-5(52), 529-535.
［2］ The international transaction processing performance council ［EB/OL］.［2010-07-01］: http: ∥www.tpc.org.
［3］ LEE S W, MOON B K, PARK C, et al. A Case for flash memory SSD in enterprise database applications ［C］ ∥ Proceedings of the ACM SIGMOD International Conference on Management of Data. Vancouver, BC, Canada: ACM, 2008: 1075-1086.
［4］ AGRAWAL D, GANESAN D, SITARAMAN R. et al. Lazyadaptive tree: An optimized index structure for flash devices ［J］. Proceeding of Very Large Databases, 2009, 1(2), 361-372.
［5］ KIM G, BEAK S, LEE H, et al. LGeDBMS: A small DBMS for embedded system with flash memory［C］ ∥ Proceedings of the 32nd International Conference on Very Large Data Bases. Seoul, Korea: ACM, 2006: 1255-1258.
［6］ LI Y N, HE B S, LUO Q, et al. Tree indexing on flash disks ［C］ ∥ Proceedings of the 25th International Conference on Data Engineering. Shanghai: IEEE, 2009: 1303-1306.
［7］ LEE S W, MOON B K. Design of flashbased DBMS: An inpage logging approach ［C］ ∥ Proceedings of the ACM SIGMOD International Conference on Management of Data. Beijing: ACM, 2007: 55-66.
［8］ TSIROGIANNIS D, HARIZOPOULOS S, SHAH M A, et al. Query processing techniques for solid state drives ［C］ ∥ Proceedings of the ACM SIGMOD International Conference on Management of Data. Providence, Rhode Island, USA: ACM, 2009: 59-72.
［9］ SHAH M A, HARIZOPOULOS S, WIENER J L, et al. Fast scans and joins using flash drives ［C］ ∥/4th Workshop on Data Management on New Hardware, DaMoN. Vancouver, BC, Canada: ACM, 2008: 17-24.
［10］ PARK S Y, JUNG D, KANG J U, et al. CFLRU: A replacement algorithm for flash memory ［C］ ∥ Proceedings of the 2006 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems. Seoul, Korea: ACM, 2006: 234-241.
［11］ KOLTSIDAS I, VIGLAS S. Flashing up the storage layer［J］. Proceeding of Very Large Databases, 2008, 1(1), 514-525.
［12］ KIM S H, JUNG D, KIM J S, et al. HeteroDrive: Reshaping the storage access pattern of OLTP workload using SSD ［EB/OL］.［2010-07-01］: http: ∥camars.kaist.ac.kr/～maeng/pubs/iwssps2009.pdf
［13］ DATAR M, GIONIS A, INDYK P, et al. Maintaining stream statistics over sliding windows ［J］. Society for Industrial and Applied Mathematics, Journal on Computing, 2002, 6(31) 1794-1813.
［14］ GIOVANNI M.S, MARIO S. Buffer management in relational databasesystems ［J］. ACM Transactions on Database Systems, 1986, 11(4), 473-498.
［15］ JOHN S B, JIRI S, STEVEN W. The diskSim simulation environment (v4.0) ［EB/OL］.［2010-07-01］: http: ∥www.pdl.cmu.edu/DiskSim.
[1] WU Yu, SHENG Zhen-hua, SHOU Li-dan, CHEN Gang. TrigSigs: an effective record linkage algorithm for unstructured data[J]. J4, 2010, 44(12): 2284-2290.
[2] SHOU Li-dan, LIAO Ding-bai, XU Chang, CHEN Gang. PWLRU: a buffer replacement algorithm for flash-based Database[J]. J4, 2010, 44(12): 2257-2262.
[3] PI Dun-Bei, CHEN Ke, CHEN Gang, DONG Jin-Xiang. Privacy protection method based on user profile of two-step sorting[J]. J4, 2010, 44(9): 1659-1665.
[4] WEI Wei, DONG E-Bei, LU Dong-Meng. Multi-resource max-min fairness and support vector machine based DDoS defense[J]. J4, 2010, 44(2): 265-270.