一种支持并发访问流的文件预取算法.pdf

ISSN 1000-9825, CODEN RUXUEW E-mail: jos@ Journal of Software , Vol.21, No.8, August 2010, pp.1820− 1833 doi: 10.3724/SP.J.1001.2010.03605 Tel/Fax: +86-10 © by Institute of Software , the Chinese Academy of Sciences . All rights reserved. ∗ 一种支持并发访问流的文件预取算法 + 吴峰光 , 奚宏生, 徐陈锋 ( 中国科学技术大学 自动化系,安徽 合肥 230027) File Prefetching Algorithm for Concurrent Streams + WU Feng-Guang , XI Hong-Sheng, XU Chen-Feng (Department of Automation, University of Science and Technology of China, Hefei 230027, China) + Corresponding author: E-mail: wfg@ Wu FG, Xi HS, Xu CF. File prefetching algorithm for concurrent streams. Journal of Software, 2010,21(8): 1820−1833. /1000-9825/3605.htm Abstract : 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. Key words: Linux; operating system; I/O performance; file prefetching; parallel I/O; access pattern 摘 要: 设计并实现了一种按需预取算法,采用更为宽松的顺序性判决条件,并以页面和页面缓存的状态作为可