一种基于预测的网格数据副本选择算法

分析影响数据副本选择的因素，将灰色系统理论应用于副本响应时间的预测，建立GM(1,1)灰色动态拟合模型；同时应用马尔可夫链预测副本可靠性发展趋势，确定系统状态转移概率矩阵，通过系统状态的划分，以概率形式分析和预测副本可靠性，并给出了两种预测算法描述。仿真实验结果表明，该算法预测效果较好,有利于作出正确的副本选择决策，并能有效实现网格中副本存储节点间的负载平衡。

Data replication selection algorithm based on prediction in data grid

A new replica selection algorithm based on prediction was proposed. After analyzing the factors infecting replica selection, the grey system theory was utilized to help predicting the data response time on the basis of GM(1,1) grey dynamic model. Meanwhile the Markov chain was applied to achieve state transition probability matrix to predict the reliability of replicas in the form of probability by the system state classification. The simulation results show that in the proposed replica selection model, the prediction is valid, helpful to selection decision, and is able to achieve load balance between nodes holding replicas.