面向5G网络云原生应用资源调度的博弈优化策略

Game Optimization Strategy of Cloud Native Application Resource Scheduling for 5G Network

随着5G网络和云原生技术的发展,面向服务的5G云原生核心网应运而生,传统应用正朝着云原生化方向发展。目前云原生服务提供商和云原生应用商数量众多且关系复杂,使得应用在云原生化过程中的资源调度面临新挑战。提出一种5G网络云原生应用资源调度优化策略,将云原生应用商和云原生服务提供商构建为多主多从的Stackelberg博弈模型,对传统收益进行具体描述并联合能耗构建利润函数和策略空间,证明给定一组微服务资源定价的情况下存在云原生应用商的纳什均衡点。在此基础上,引入柯西分布对策略进行优化,提高其收敛性能,通过分布式迭代方法得到云原生服务提供商的最佳微服务定价和云原生应用商的最佳微服务租用比例。仿真结果表明,相比ACA算法、QOS PA算法以及GOS策略,该策略能够有效提高网络收益和用户体验质量,同时降低应用开发能耗。

The development of 5G network and cloud native technologies has led to the emergence of service-oriented 5G cloud native core networks,and traditional applications are also developing in the cloud native direction.However,the large number of cloud native service and application providers and their complex relationships impose a new challenge to the resource scheduling of applications in cloud native development.To address the problem,this paper proposes a game optimization strategy of cloud native application resource scheduling for 5G network.The cloud native service and application providers are built as a multi-leader and multi-follower Stackelberg game model to specify the traditional revenue,which is combined with the energy consumption to build a profit function and strategy space.On this basis,it is proven that there is a Nash equilibrium point for cloud native application providers when a set of microservice resource pricing is given.Then the Cauchy distribution is introduced to optimize the strategy and improve its convergence performance.Through a distributed iterative method,the best microservice price and the optimal microservice lease ratio of cloud native application providers are obtained.The simulation results show that compared with the ACA algorithm,QOS PA algorithm and GOS strategy,the proposed strategy can effectively improve network revenue and user experience quality,and reduce energy consumption for application development.