File prefetching algorithm for concurrent streams

被引：2
|
作者：
Wu F.-G. [1 ]
Xi H.-S. [1 ]
Xu C.-F. [1 ]
机构：
[1] Department of Automation, University of Science and Technology of China
来源：
Ruan Jian Xue Bao/Journal of Software | 2010年 / 21卷 / 08期
关键词：
Access pattern; File prefetching; I/O performance; Linux; Operating system; Parallel I/O;
D O I：
10.3724/SP.J.1001.2010.03605
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
This paper proposes and implements a demand prefetching algorithm, which uses relaxed sequentiality criteria as well as page and page cache states as reliable sources of information. It can discover and prefetch sequential streams mixed in random accesses. It can also support the interleaved access patterns created by concurrent sequential reads on one file descriptor. Experimental results show that it performs much better than legacy Linux readahead: sequential reading intermixed with random ones are faster by 29%; I/O throughput of interleaved streams could be 4~27 times better, with application visible I/O latencies improved by up to 35 times. This demand prefetching algorithm has been adopted by Linux kernel 2.6.24. © by Institute of Software, the Chinese Academy of Sciences. All rights reserved.
引用
收藏
页码：1820 / 1833
页数：13
相关论文
共 24 条
  • [1] Shriver E., Small C., Smith K.A., Why does file system prefetching work?, Proc. of the Annual Technical Conf. on 1999 USENIX Annual Technical Conf. (ATEC'99), pp. 71-84, (1999)
  • [2] Schmid P., 15 years of hard drive history: Capacities outran performance, (2006)
  • [3] Kroeger T.M., Long D.D.E., Design and implementation of a predictive file prefetching algorithm, Proc. of the General Track: 2002 USENIX Annual Technical Conf., pp. 105-118, (2001)
  • [4] Lei H., Duchamp D., An analytical approach to file prefetching, Proc. of the 1997 USENIX Annual Technical Conf., pp. 275-288, (1997)
  • [5] Paris J.F., Amer A., Long D.D.E., A stochastic approach to file access prediction, Proc. of the Int'l Workshop on Storage Network Architecture and Parallel I/Os, pp. 36-40, (2003)
  • [6] Whittle G.A.S., Paris J.F., Amer A., Long D.D.E., Burns R., Using multiple predictors to improve the accuracy of file access predictions, Proc. of the 20th IEEE/11th NASA Goddard Conf. on Mass Storage Systems and Technologies (MSS 2003), pp. 230-240, (2003)
  • [7] Li Z.M., Chen Z.F., Zhou Y.Y., Mining block correlations to improve storage performance, ACM Trans. on Storage, 1, 2, pp. 213-245, (2005)
  • [8] Cao P., Felten E.W., Karlin A.R., Li K., Implementation and performance of integrated application-controlled file caching, prefetching, and disk scheduling, Trans. on Computer Systems, 14, 4, pp. 311-343, (1996)
  • [9] Patterson R.H., Gibson G.A., Ginting E., Stodolsky D., Zelenka J., Informed prefetching and caching, Proc. of the 15th ACM Symp. on Operating Systems Principles, pp. 79-95, (1995)
  • [10] Patterson R.H., Gibson G.A., Satyanarayanan M., A status report on research in transparent informed prefetching, ACM SIGOPS Operating Systems Review, 27, 2, pp. 21-34, (1993)
123