离线浏览下载引擎算法的优化及改进  

Optimization and Refinement of the Algorithm of an Offline Browser Download Engine

作　　者：吴晓[1] 刘君悦[1] 孙玉芳[1] 李彤[1] 

机构地区：[1]中国科学院软件研究所,北京100080

出　　处：《计算机科学》2002年第12期103-105,117,共4页Computer Science

基　　金：国家自然科学基金(60073022); 国家863高科技项目(863-306-ZD12-14-2); 中科院知识创新基金(KGCXI-09)

摘　　要：The refined algorithm is put forward mainly based on the Wget. Combining the advantages of HTTrackand Wget, the download algorithm is rewritten on the following aspects: the download process is changed fromdepth-first retrieval to width-first retrieval; from single-socket to multi-socket; and from recursive retrieval to loopretrieval. The original process,which includes the system calls that maybe cause the network blocking,is cutted intodifferent sectors,and runs in the way of multitask. The results of experiments and feedbacks from actual applicationshow that the refined algorithm gets great improvement on the download efficiency and memory occupancy.The refined algorithm is put forward mainly based on the Wget. Combining the advantages of HTTrack and Wget, the download algorithm is rewritten on the following aspects: the download process is changed from depth-first retrieval to width-first retrieval; from single-socket to multi-socket; and from recursive retrieval to loop retrieval. The original process,which includes the system calls that maybe cause the network blocking, is cutted into different sectors,and runs in the way of multitask. The results of experiments and feedbacks from actual application show that the refined algorithm gets great improvement on the download efficiency and memory occupancy.

关 键 词：计算机网络 局域网 INTERNET 服务器 离线浏览下载引擎算法 优化 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]