基于云理论的网格资源选择  

Grid resource selecting based on cloud theory

作　　者：马满福[1,2] 段富海 黄志毅 章卫国[1] 

机构地区：[1]西北工业大学自动化学院,西安710072 [2]兰州飞行控制有限责任公司航技中心,兰州730070

出　　处：《计算机应用》2009年第4期1162-1164,共3页journal of Computer Applications

基　　金：教育部科学技术研究重点项目(208148);甘肃省科技攻关项目(2GS064-A52-035-03)

摘　　要：资源选择是影响网格调度和系统效率的关键,针对网格资源选择中用户对服务质量(QoS)的定性描述和调度的自私性,提出了利用云理论实现资源选择的方法。在深入分析QoS参数的云理论模型基础上,提出了以资源代理实现云模型资源选择的体系结构,设计了相应的调度算法。实验表明,该算法在资源调度率和吞吐量以及系统资源的利用效率等方面体现出良好的特性,同时克服了用户定义QoS参数的困难,达到了优化调度的目的。Resource selecting is the key for scheduling and system efficiency in grid. Concerning the qualitative description of Quality of Service （QoS） by grid users and selfishness of scheduling, grid resource selecting based on cloud theory was proposed. QoS parameters based on the cloud model were analyzed, as the requirement; the author proposed a new grid resource broker architecture. At last, the optimized scheduling algorithm was described based on the model. Finally the author performed simulations to compare the performance of optimized scheduling algorithm with that of the time minimization and cost minimization. The experimental results show that the optimized scheduling algorithm is efficient on throughput and system utilization ratio within grid computing environments.

关 键 词：网格 云模型 代理体系结构 调度算法 

分 类 号：TP302[自动化与计算机技术—计算机系统结构]