摘要
数据网格中由于采用复制备份策略,文件存在多个副本。用户访问文件时,如何从拥有相同文件的多个结点中选择某一结点获取文件,达到以相同的代价获得最优质的服务,是一个迫切需要研究的问题。深入研究了蚂蚁算法的原理,分析了影响副本选择性能的主要因素,利用这些参考因素设计了基于蚂蚁算法的副本选择策略;并对这个新算法进行了分析和实现。经仿真平台实验,表明该算法可有效地减少数据访问延迟及带宽消耗,实现网格中存储节点间的负载平衡,提高数据的访问速度。
There are replication of data in data grid. But how to select a site which have the requested data to fufill the user＇ s demand hasn＇t been solved up to the present. And a user should get the best services with the least cost. This problem is a multiobjective pro- gramming problem. The replica selection strategy is designed after making a deep research into the ant algorithm and investigating the key factors that influence the performance of replica selection greatly. This strategy is analyzed and implemented. Through using simulator, test results that this new ant algorithm can reduce data access latency, decrease bandwidth consumption and distribute storage site load, improve data access speed.
出处
《计算机工程与设计》
CSCD
北大核心
2008年第23期6157-6160,共4页
Computer Engineering and Design
基金
重庆市教育委员会科学技术研究基金项目(KJ071203)
重庆文理学院重点科研项目(Z2006sj31)
重庆文理学院科研启动经费课题基金项目(2007036)。
关键词
网格计算
数据网格
蚂蚁算法
副本选择
grid computing
data grid
ant algorithm
replica selection
