面向不确定需求租户集的云服务智能匹配模型

为了解决云计算环境下由海量租户集和资源集间的不确定性因素引起的高质量云服务获取困难的问题,提出了一种描述动态异构租户集不确定性需求的方法.在此基础上,构建属性权重完全未知情况下的云服务智能匹配模型,排除了租户提交权值造成的偏差.神经网络以属性区间计算的相离度作为输入,服务满意度为输出来动态模拟租户集的不确定需求,运用萤火虫算法求解模型获取最优服务组合.最后,实例验证了神经网络的可靠性以及算法的有效性.实验结果表明,模型能有效获取高质量的云服务组合,优于传统的匹配方法.

Multi-tenant Sets Oriented Cloud Service Composition Intelligence Matching Model with Uncertain Demands

In order to solve the difficult high-quality cloud service matching problem caused by the uncertain factors between the massive tenants set and resource set,a method to describe the uncertain demand of dynamic and heterogeneous tenants set is proposed.And then a cloud service intelligence matching model,in which deviation caused by the weight submitted by tenants is excluded,is built under the circumstance of the attribute weights is unknown completely.The neural networks dynamically simulates the tenants set's uncertain demand by using the deviation degree calculated by attribute interval numbers as the input and the service satisfaction as the output,the optimal combinations of cloud service are obtained by using FA algorithm to solve the model.Finally,a validation instance is conducted to demonstrate the reliability of neural networks and the effectiveness of the algorithm..The experimental result shows that the model can effectively obtain the high-quality cloud service composition,which is better than the traditional service matching methods.