基于PSO改进算法的气象数据网格任务调度

Meteorological Data Grid Task Schedule Based on PSO Improved Algorithm

为提高在有限带宽下气象观测中心海量数据的任务调度和数据传输效率,提出一种基于粒子群优化(PSO)改进算法的气象数据网格任务调度算法。给出副本域的概念,将PSO算法与副本域相结合,设计任务调度模型和符合气象数据网格环境的目标函数。仿真结果表明,该算法完成调度的时间小于遗传算法和穷尽搜索算法,收敛速度快于离散型PSO算法,且更加稳定。

In order to improve the efficiency of task schedule and data transmission about the massive data of weather bureau under limited bandwidth, this paper proposes a meteorological data grid task schedule algorithm based on Particle Swarm Optimization（PSO） improved algorithm. It gives the conception of Replica Domain（RD）, makes combination of PSO algorithm, and designs task schedule model and the objective functions which conform to the meteorological data grid environment. Simulation results show that the finishing scheduling time of this algorithm is less than Genetic Algorithm（GA） and end search algorithm, its convergence speed is faster than Discrete Particle Swarm Optimization（DPSO） algorithm, and is more stable.