摘要
在传统基于层次模型的网格体系中引入了基于价格的市场机制,构建了一种基于经济模型的网格体系结构。网格用户在关注时间和费用的同时,也应注重资源的使用价值。本文构造了网格资源的效用函数和Cobb-Douglas形式的用户效用函数,用以表达网格用户不同的偏好和目标,并利用Lagrange方法对效用函数进行优化,给出了基于效用函数优化的资源分配与动态调价策略。最后进行了仿真实验,验证本文所给出的网格任务—资源映射算法的性能和效率。
In this paper,the market mechanism based on price is imported to the traditional Grid architecture based on hierarchical model,and a Grid architecture based on economic model is constructed. Grid users always concern about the time and cost as well as the use value of grid resources. Accordingly,a utility function for grid resource is constructed as well as a grid user utility function having the Cobb-Douglas form,which can express the different preferences and objectives. The Lagrange method is used to solve the optimal problem of utility function. In addition, a strategy for resource allocation and pricing based on optimization of utility function is proposed. Finally, performance and efficiency of the grid tasks and resources mapping algorithm proposed in this paper is analyzed through the simulation experiment.
出处
《微计算机信息》
2009年第36期156-158,共3页
Control & Automation
关键词
网格计算
经济模型
效用函数
资源分配
动态调价
grid computing
economic model
utility function
resource allocation
dynamic price adjustment