仿真网格中服务的最优化调度机制研究  

Research on the Optimization Scheduling Mechanism for Service Within Simulation Grid

作　　者：邝坪[1] 金海[1] 袁平鹏[1] 

机构地区：[1]服务计算技术与系统教育部重点实验室集群与网格计算湖北省重点实验室华中科技大学计算机学院,湖北武汉430074

出　　处：《小型微型计算机系统》2009年第1期8-12,共5页Journal of Chinese Computer Systems

基　　金：国家自然科学基金重大专项(90412010)资助

摘　　要：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.

关 键 词：仿真网格 服务 调度 遗传模拟退火算法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]