节点文献

基于启发式算法的数据迁移策略

Cloud Data Migration Policy Based on Heuristic Algorithm

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 孙阳许建潮郭建伟刘钢

【Author】 SUN Yang;XU Jian-chao;GUO Jian-wei;LIU Gang;College of Computer Science and Engineering,Changchun University of Technology;

【机构】 长春工业大学计算机科学与工程学院

【摘要】 随着云计算和大数据的飞速发展,如何高效地维护和管理云数据中心已经成为了业界关注的主要问题.通过对数据迁移策略的深入研究,提出了一个基于启发式算法的数据迁移策略IB-FA.主要思想包含两部分:首先基于萤火虫算法根据各个物理服务器的负载情况进行全局搜索,为避免个别萤火虫陷入局部最优引入了自适应惯性权重;其次,通过目标函数约束限制所求解,使其在负载均衡及带宽利用率方面都更加优化.实验结果表明:与现有的迁移策略对比,IB-FA能找到一个更适合数据的迁移位置,在分散I/O操作的同时不仅优化了数据访问速度,而且提高了服务器资源利用率和带宽利用率.

【Abstract】 With the rapid development of cloud computing and big data,how to effectively maintain and manage the cloud data center has become a major concern in the industry. A data migration strategy based on heuristic algorithm IB- FA is proposed after the research of the data migration strategy. The main idea includes two parts: first of all,based on the heuristic algorithm IB- FA is designed by the load of each physical server to global search according to the objective function and introduced the adaptive inertia weight to avoid trapping in local optimum. Secondly,constrained by the objective function the solution was more optimized in the load balancing and bandwidth utilization. The experimental results showed that compared with the existing migration policies, the IB- FA algorithm found a more suitable location for the target migration data. The data access speed was optimized while the I / O operation was dispersed and the utilization rate of server resource and bandwidth utilization were also improved.

【基金】 吉林省自然科学基金项目(20101533);吉林省教育厅“十二五”科学技术研究项目(吉教科合【2011】第103号)
  • 【文献出处】 吉林建筑大学学报 ,Journal of Jilin Jianzhu University , 编辑部邮箱 ,2016年03期
  • 【分类号】TP301.6
  • 【下载频次】61
节点文献中: 

本文链接的文献网络图示:

本文的引文网络