在网格中,如何为任务提供最优化资源服务是一个十分复杂的问题.本文以仿真网格为基础,建立一种服务的最优化调度的理论模型.根据这种理论模型,提出一种新的遗传模拟退火算法.从而形成一种仿真网格环境下服务的最优化调度机制.为了验证这种机制的可行性和有效性,开发一个基于此服务调度机制的仿真网格运行管理系统.通过这个系统的仿真实验表明:这种服务的最优化调度机制具有良好的可行性和有效性.
In Grid, it is very difficult that how to provide the optimization services for tasks. In this paper, a theoretical scheduler model of optimization service is presented for the material tasks and the services. Fgr solving the theoretical model, we present a new genetic simulated annealing algorithm. All these constitute the scheduler mechanism of optimization service for task within simulation Grid. In order to validate the feasibility and the availability of this scheduler mechanism, we develop a service scheduling management system about simulation Grid. By using the service scheduling management system, a series of simulation experiments are done. The results show that the scheduler mechanism of optimization service is feasible and efficient within simulation Grid.