中文摘要：

为满足物联网终端结点移动性和数据访问实时性要求,根据云存储中副本分布与访问的模式,提出结合数据访问点位置和访问频率,具有低访问代价的存储服务器选取方法。考虑负载对访问代价的影响,通过动态更新代价矩阵,提出负载均衡机制。研究结果表明：该方法在最少副本数目的前提下,将副本分配到与访问情况相适应的具有较低访问代价的存储服务器上,可有效减少数据访问开销,并避免副本集中分布,保证各存储服务器的负载均衡,更适合于物联网环境中云存储数据管理。

英文摘要：

In order to meet the requirements of mobile terminal in terms of mobility and real time in the Internet of Things,according to the model of replica distribution and data access in cloud storage,a method for selecting low access cost storage server was presented on the basis of access positions and frequencies.Considering the effect of the load on access cost,the load-balancing technique by updating cost matrix dynamically was presented.The results show that when achieving the minimum number of replicas,the replicas are adaptively distributed to the storage servers according to the data access status with low access cost.With reduced accessing cost,avoided replicas non-equalizing distribution and enhanced load balance,the method is favorable in management of cloud storage in the Internet of Things.