博弈模型下云资源分配纳什均衡算法（英文）

Nash equilibrium algorithm of cloud resource allocation under game model

为解决云资源分配中用户的异构性需求问题,以经济模型为基础,运用非合作博弈论研究了具有完全理性的云资源分配问题,提出基于非合作博弈的资源分配纳什均衡算法,该算法以博弈效用最优化为目标,利用正比例共享机制按用户出价分配云资源。求解了用户的出价函数,重点证明了出价函数的凹性以及多用户间的最优出价组合的纳什均衡存在性与唯一性,并提出了资源分配纳什均衡算法。实验结果表明:非合作博弈云资源分配算法不仅能够按市场变化反映资源价格与用户需求之间的浮动关系,从而规范用户间的资源公平竞争关系,而且可以使资源分配更加合理和均衡。

Based on the economic model,this paper studied cloud resource allocation problem with complete reality by using non-cooperative game theory to solve the problem of heterogeneity of users’requirements in cloud resource allocation,and a resource allocation Nash equilibrium algorithm based on non-cooperative game theory is proposed.Take the game utility optimization as an objective,the algorithm allocates cloud resource according to user’s bidding by direct proportional sharing mechanism.The bidding function of cloud users is solved,more important,the concavity of the bidding function is proved.Meanwhile,we prove the existence and uniqueness of Nash equilibrium solution of optimal bidding set among multiple users.Then,a resource allocation algorithm solving Nash equilibrium solution is designed.The experimental results show that the non-cooperative game resource allocation algorithm can reflect the relationship between users’requirement and resource’s price according to the market change,which can not only regulate the fair resource competing relationship,but also make resource allocation more fair and more equilibrium.