基于PSODE的网格任务调度算法

针对网格计算中任务在各个资源之间的调度问题,提出了一种网格环境下PSODE的任务调度算法.该算法实现了计算资源、存储资源、带宽资源、数据资源的利用率最高化和代价最低化.对基本粒子群算法和差分进化算法进行了分析,通过构造算法函数、适应值函数和权重公式,建立了粒子群差分混合算法并对其进行优化,介绍了算法的实现过程.实验结果表明,该算法与其它调度算法比较,具有良好的性能.

Grid job scheduling algorithm based on PSODE

A kind of task scheduling algorithms PSODE(particle swarm optimization’s differential evolution algorithm),is proposed in grid environment for the problem of task scheduling among the various resources in grid computing.The algorithm has realized maximum utilization and cost minimization of computing resources,storage resources,bandwidth resources as well as data resources.First of all,particle swarm optimization and differential evolution algorithm are analyzed,then particle swarm’s differential hybrid algorithm is established and its optimization by constructing the arithmetic functions,the fitness function and weight formula.Finally the algorithm’s implementation process is introduced.Through experimentation,it is proved that the algorithm has good performance compared with other scheduling algorithms.