基于随机博弈的云代理与私有云联盟的最优收益

提高私有云联盟（Federated Private Clouds）中虚拟机利用率问题是实现云代理（Cloud Broker）和私有云联盟最优收益的关键，引入随机博弈算法可分别使双方达到最优收益。首先，分析云代理和私有云联盟任务到达的随机特性，利用随机博弈模型解决虚拟机租用数量的收益获得问题；其次，分析并确定该模型为变和随机博弈，提出了反向迭代算法实现均衡；最后；仿真实验采用亚马逊EC2的价格，与租用固定虚拟机数量的方法进行对比，结果表明采用该方法能够使博弈双方都达到较高收益。

Optimal Revenue of Cloud Broker and Private Cloud Federation based on Stochastic Games

Improving the utilization of virtual machines in Federated Private Clouds(FPC) is key to achieving the optimal benefits of Cloud Broker(CB) and Federated Private Clouds(FPC). Firstly, the random characteristics of CB and FPC are analyzed, and the stochastic game model is used to solve the problem of virtual machine rental; Secondly, we analyze and determine the model as changeable sum stochastic game, and propose a reverse iterative algorithm to achieve equilibrium. Finally, using the price of Amazon EC2 to compare with the method of renting the fixed virtual machine, the simulation shows that this method can make both sides of the game achieve higher revenues.