摘要
针对云数据中心资源利用率低、云服务提供商收益低等问题,提出一种基于生产函数的云服务提供商收益最大化同时兼顾用户满意度的资源调度算法。该算法将资源调度分两阶段处理,首先合理规划云服务器所有资源,最优化配置资源;然后结合用户请求,云服务代理从资源池选择配置好的资源并分配资源给用户,通过两阶段的算法实现,解决了云数据中心资源利用率低、云服务提供商收益低等问题。通过与基于博弈的效用优化算法进行比较,仿真结果表明该调度算法具有更好的性能。
In order to address the issue of low resource utilization and low cloud service provider gains on cloud data center, this paper presented a resource scheduling strategy, which not only maximized the profit of cloud service providers, but also achieved the highest customer satisfaction based on production function. There were two stages in this scheduling solutions. Firstly, it rational planned all resources in cloud data center to optimized allocation resources. Then, it combined with users' request. Cloud server agents allocated resources to cloud users. After achieve this two stages' algorithm, it solved the problem of tow resource utilization and low cloud service provider gains in cloud data center. At last, it compared with the effectiveness of optimization algorithm based on game. Simulation experimental results show that the proposed method has a better performance.
出处
《计算机应用研究》
CSCD
北大核心
2017年第2期397-400,452,共5页
Application Research of Computers
基金
长江学者和创新团队发展计划资助项目(IRT1299)
国家"863"计划资助项目(2015AA01C303)
关键词
云计算
资源调度
生产函数
用户效用
cloud computing
resource scheduling
production function
users' satisfaction