摘要
在计算机网格环境中,用户对资源的管理策略以及存取花费的代价是不相同的,而且难以用实验测试各种任务调度方法的有效性。本文在分析了目前已有的经济学模型资源分配框架基础上,提出了一个能满足用户各种服务质量(Quality of Senrice,QoS)要求的作业管理和调度系统,并设计和实现了基于QOS的网格作业管理系统和基于遗传算法的任务调度策略。
Every organization controlling over resource tactics and accessing to resource spend has nothing in common with each other in a grid environment. It is difficult to test various kinds of validity for different tasks scheduling methods with the experiment. In this paper, the existing economics model resource allocation framework is analyzed, which meets various kinds of job managements and job schedule of QoS＇s request of users. The design and realization of the job management system based on QoS and strategy of task scheduling based on GA（genetic algorithm. ）are described in detail.
出处
《国外电子测量技术》
2008年第11期33-35,69,共4页
Foreign Electronic Measurement Technology
基金
国家863高技术研究发展计划项目高性能计算机及其核心软件-西安网格结点建设(编号:2002AA04550)
中国教育科研网格项目(编号:CG2003-CG008)
