服务于凤电系统的改进缓存替换算法研究-计算机科学.pdfVIP

Vo1. 44 No. 9 第 44 卷第9 期 计算机科学 2017 年 9 月 COMPUTER SCIENCE Sep.2017 服务于凤电系统的改进缓存替换算法研究 鲁尔洁陈峦李坚黄琦张真源井实周统汉 (电子科技大学能源科学与工程学院 成都 611731) 摘 要针对风电系统中缓存命中率较低等问题，在最近最少使用算法(Least Recently Used , LRU) 、最不经常访问 算法(Least Frequently Used， LFU) 、 SIZE 以及 Hybrid 算法的基础上，提出了一种基于综合因素的替换算法 FST (Frequency ,Object Size ,Access Time) ，从而解决了传统算法考虑因素羊一、系统性能较低等问题。该算法结合了访 问频率、对象大小、访问时间间隔及最久未访问等特性，并根据最近访问时间长短采取分段的处理方法。在风电系统 的缓存服务器中，将 FST 算法与 LRU， LFU 和 SIZE 算法进行实验对比，实验结果显示 FST 算法在提高命中率、减少 延迟时间方面具有更好的性能。 关键词 风电系统，综合因素，替换算法，命中率，延迟时间 中固法分类号 TP393 文献标识码 A DOI 10. 11896/j. issn. 1002-137X. 2017. 09. 043 Research on Improved Cache Replacement Algorithm Serving for Wind Power System LU Er-jie CHEN Luan LI Jian HUANG Qi ZHANG Zhen-yuan J卧~G Shi ZHOU Tong-han (School of Energy Science and Engineering ,University of Electronic Science and Technology of China ,Chengdu 611731 ,China) Abstract Airning at the problems such as the low cache hit rate in wind power system , after researching on LRU (Least Recently Used) ,LFU CLeast Frequently Used) , SIZE and Hybrid , a replacement algorithm FST (Frequency , Object Size ,Access Time) based on comprehensive factors was proposed to solve the problems such as single factor and low system s performance which are caused by traditional algorithms. This algorithm combines the features of access frequency ,0 bj ect si况， access time interval and the longest time without access , and it takes the method of segmentation according to the length of the recent access time. By taking contrast test with LRU ,LFU and SIZE in the wind power system cache server ,FST algorithm shows better performance in improving the hit rate and reducing the delay time