信任驱动机制下的网格负载调度算法

Grid Schedule Algorithm Based on Load Under Trust-driven Mechanism

利用网格信任模型与效益函数,结合资源当前负载状况,提出一种基于负载均衡的任务调度算法——Trust-Driven_Load(TD_Load),在满足最大信任效益值的条件下,采用预计最短完成时间对多个资源进行选择。实验结果表明,在相同的条件设置下,TD_Load算法在资源负载、makespan和任务平均等待时间上优于基于信任效益值的传统算法,而且算法时间花费小,当任务数量增多时,综合调度性能更优。

Under trust-driven mechanism,associating trust model,benefit function with current load of resources,a new job schedule algorithm TD_Load based on load balancing is presented for computing grid.Under the condition of the best trust benefit value,the schedule chooses resources by MCT calculated in advance.Under the same conditions,experimental results show that TD_Load algorithm is better than traditional algorithms based on benefit values in resource load,makespan,and average waiting time.At the same time,the cost of the algorithm itself is not much,so the total schedule performance is excellent,especially when the number of jobs increases.