Game strategies among multiple cloud computing platforms for non-cooperative competing assignment user tasks

With the development of cloud computing technology, more and more cloud computing platforms are emerging which provide a great deal of computing services. But every cloud computing platform is selfish, there are many competitions and contradictions among them. Only if many cloud computing platforms are united and cooperative, they can realize resource sharing and aggregate a greater capacity. Therefore, this paper studies game strategies among multiple cloud computing platforms for coordinating and competing assignment user tasks. Firstly, we present a framework of federated cloud computing platforms, which is composed of multiple private clouds. User tasks will be received, assigned, and executed by the federated cloud. Secondly, a non-cooperative game model for tasks allocation is established among multiple private clouds. Then, the Nash equilibrium solution under the non-cooperative game is transformed into a function optimization problem. We use the particle swarm optimization algorithm to obtain near optimal solution. Finally, based on the Nash equilibrium solution, we propose a tasks allocation algorithm called NGTA for multiple private clouds in the non-cooperative game. The experimental results show that in the presence of competition, compared with Max–Min and Min-Min algorithm, NGTA algorithm can bring balanced utility satisfaction to each private cloud. The expected utility deviation of each private cloud is very small, the average is about 0.002.