期刊文献+

基于ARC的闪存数据库缓冲区算法

Buffer Replacement Algorithm for Flash-Based Databases Based on ARC
下载PDF
导出
摘要 闪存是一种纯电子设备,具备体积小、数据读取速度快、能耗低、抗震性强等优点,被用来部分替代机械硬盘从而提升存储系统的性能.但是,现有的缓冲区置换算法都是针对机械硬盘的物理特性进行设计和优化,因此有必要针对闪存的物理特性重新设计缓冲区置换算法.提出一种新的面向闪存数据库的缓冲区替换算法CF-ARC.算法设计了一种新的页替换机制,即在替换干净页或者脏页的时候考虑其访问频度的大小,优先将访问频度少的干净页替换出缓冲区,使得热页继续留在缓冲区提高命中率,从而获得更好的性能,通过对实验结果的对比分析发现CF-ARC在多数情况下具有比其它置换算法更高的性能. Flash memory is a pure electronic equipment and has the advantages of smaller volume,faster reading speed,lower power consumption and strong vibration resistance,so it is used to partly replace the disk to improve the performance of storage system.But the existing design and optimization of buffer replacement algorithms are based on the physical characteristics of mechanical hard disk.Therefore,it is necessary to redesign a new buffer replacement algorithm which contrapose the physical characteristics of flash memory.This study presents a new buffer replacement algorithm named CF-ARC.A new type of page mechanism replacement is designed,which means the access frequency should be considered when the clean or dirty pages are replaced.The clean pages less visit should replace the buffer to improve the hit rate in hotspot and achieve a better performance.The experimental results show that CF-ARC has better performance than other buffer replacement algorithms in most cases.
出处 《计算机系统应用》 2018年第3期156-161,共6页 Computer Systems & Applications
基金 国家自然科学基金(61502102 61402109 61370078)
关键词 闪存数据库 缓冲区置换算法 替换机制法 传统机械式硬盘 ARC算法 flash databases buffer replacement algorithm replacement menchanism traditional mechanical disk ARC algorithm
  • 相关文献

参考文献1

二级参考文献59

  • 1Chang L P, Kuo T W. Efficient management for large-scale flash memory storage with resource conservation. IEEE Transactions on Storage, 2005, 1(4): 381-418.
  • 2Wu C H, Kuo T W. An adaptive two level management for the flash translation layer in embedded systems//Proeeedings of the International Conference on Computer Aided Design (ICCAD'06). San Jose, USA, 2006 : 601-606.
  • 3Gray J. Tape is dead, disk is tape, flash is disk, RAM locali ty is king//Proceedings of the Gong Show Presentation at Third Biennial Conference on Innovative Data Systems Research: Pacific Grove, USA, 2007:1-12.
  • 4Lee S W, Moon B. Design of flash-based DBMS: An in page logging approach//Proceedings of the ACM SIGMOD Inter national Conference. Beiiing, China, 2007:55-66.
  • 5Canim M, Bhattacharjee B, Mihaila G A, et al. An object placement advisor for DB2 using solid state storage. Proceed ings of the VLDB Endowment, 2009, 2(2): 1318-1329.
  • 6LeeSW, MoonB, ParkC, KimJ M, KimSW. Aeasefor flash memory SSD in enterprise database applications// Proceedings of the ACM SIGMOD International Conference. Vancouver, Canada, 2008:1075-1086.
  • 7O'Neil E J, O'Neil P E, Weikum G. The LRU-K page replacement algorithm for database disk buffering//Proceedings of ACM SIGMOD International Conference. Washington, USA, 1993:297-306.
  • 8Jiang S, Zhang X D. LIRS: An efficient low inter reference recency set replacement policy to improve buffer cache per formance//Proceedings of the ACM SIGMETRICS Confer ence on Measurements and Modeling of Computer Systems. California, USA, 2002:31-42.
  • 9Park S Y, Jung D, Kang J, Kim J S. CFLRU: A replace- ment algorithm for flash-memory//Proceedings of the Inter- national Conference on Compilers, Architecture and Synthe- sis for Embedded Systems. Seoul, Korea, 2006:234-241.
  • 10Jung H, Shim H, Park S, et al. LRU-WSR: Integration of LRU and writes sequence reordering for flash memory. IEEE Transactions on Consumer Electronics, 2008, 54(3): 1215- 1223.

共引文献25

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部