启发式云计算多源资源访问特征最小方差估计

Minimum Variance Estimation of Multi Resource Access Feature Based on Heuristic Cloud Computing

提出一种基于启发式云计算的多源资源访问特征最小方差估计算法,构建多源资源访问的云计算Cloud-P2P融合模型,采用遗传算法对Cloud-P2P融合模型中多源资源访问特征进行信息提取,给出多源资源信息访问特征状态,采用自适应全局概率搜索,自主对整个所需搜索范围进行搜索以及优化,实现对多源资源访问特征的最小方差估计,提高访问性能。仿真结果表明,算法特征信息提取精度高,方差估计准确,通过启发式云计算对多源资源信息系统访问特征的最小方差准确估计,利用了众多闲置的普通用户终端节点上蕴含的巨大的计算和存储资源,可灵活设置备份规模,鲁棒性高,性能优越。

A calculation based on the heuristic cloud multi-source resource access characteristics of minimum variance estimation algorithm is proposed, building the multiple resource access cloud computing Cloud-P2 P fusion model, genetic algorithm is used in Cloud-P2 P fusion model of information extraction features multi-source access resources, given the multisource information resources access feature state, using adaptive global probability search, independent of the whole required the scope of the search for search and optimization, to achieve the minimum variance access characteristics of multisource resource estimation, improve the access of. The simulation results show that, the algorithm can extract the feature information of high precision, variance estimation accuracy, minimum variance over heuristic cloud computing access characteristics of multi-source information resource system accurate estimation, using the common user terminal node contains many idle on the huge computing and storage resources, it can flexibly set the backup size, it has high robustness and superior performance.