基于Stackelberg博弈的云CDN缓存资源分配算法
发布时间：2018-07-22 15:24
【摘要】：为提高云存储环境下内容分发网络(CDN)的分发效率,提出一种基于Stackelberg博弈的缓存资源分配与定价算法。将Web服务器和云CDN代理商建模成一个多主多从的Stackelberg博弈模型,并构建其各自的效用函数。证明在代理商价格确定的情况下存在Web服务器纳什均衡点,利用一种分布式迭代算法求解博弈模型,得到最优定价与该定价下的最优缓存分配结果。仿真结果表明,该算法可保证Web服务器缓存需求在代理商之间的高效分配,与用户服务质量优先算法相比,可使Web服务器获得更高的单位成本效益。
[Abstract]:In order to improve the efficiency of content distribution network (CDN) in cloud storage environment, a buffer resource allocation and pricing algorithm based on Stackelberg game is proposed. Web servers and cloud CDN agents are modeled as a multi-principal multi-slave Stackelberg game model and their respective utility functions are constructed. It is proved that there exists a Web server Nash equilibrium point when the agent price is determined. A distributed iterative algorithm is used to solve the game model, and the optimal pricing and the optimal cache allocation under this pricing are obtained. The simulation results show that the proposed algorithm can ensure the efficient allocation of cache requirements among agents, and it can achieve higher unit cost benefit than the QoS priority algorithm.
【作者单位】： 北京信息科技大学信息与通信工程学院;
【基金】：国家自然科学基金“基于资源标签交换的无线网络端到端能效管理策略研究”(61271198) 北京市教委科技计划重点项目“分层合作博弈的端到端绿色网络体系与关键技术研究”(KZ201511232036)
【分类号】：TP393.0
，
本文编号：2137897
[Abstract]:In order to improve the efficiency of content distribution network (CDN) in cloud storage environment, a buffer resource allocation and pricing algorithm based on Stackelberg game is proposed. Web servers and cloud CDN agents are modeled as a multi-principal multi-slave Stackelberg game model and their respective utility functions are constructed. It is proved that there exists a Web server Nash equilibrium point when the agent price is determined. A distributed iterative algorithm is used to solve the game model, and the optimal pricing and the optimal cache allocation under this pricing are obtained. The simulation results show that the proposed algorithm can ensure the efficient allocation of cache requirements among agents, and it can achieve higher unit cost benefit than the QoS priority algorithm.
【作者单位】： 北京信息科技大学信息与通信工程学院;
【基金】：国家自然科学基金“基于资源标签交换的无线网络端到端能效管理策略研究”(61271198) 北京市教委科技计划重点项目“分层合作博弈的端到端绿色网络体系与关键技术研究”(KZ201511232036)
【分类号】：TP393.0
，
本文编号：2137897
本文链接：https://www.wllwen.com/guanlilunwen/ydhl/2137897.html
