期刊文献+

开源IaaS平台中的随机调度任务动态融合仿真

Open Source IaaS Platform of Random Scheduling Tasks Dynamic Simulation
下载PDF
导出
摘要 由于开源IaaS平台中用户数量众多,系统所需处理的任务量十分巨大,导致传统的粒子群调度方法,容易陷入局部最优,求解到一定范围时会存在大量冗余迭代,处理效率低等问题。为提高资源利用率,提出一种采用遗传和蚁群动态融合的开源IaaS平台中随机调度任务方法,首先分析了蚁群算法信息素初始化、选择路径、更新信息素的详细过程。其次介绍了遗传算法染色体编码、初始种群的产生、选择算子、交叉算子和变异算子的具体步骤。然后给出开源IaaS平台中随机任务调度目标,求解出开源IaaS平台中各资源完成该资源中全部子任务所需的时间。最后通过遗传算法准确形成开源IaaS平台任务调度的初始解,并以初始化为信息素分布,实现依据蚁群算法正反馈和高效收敛的优点获取最佳随机任务的调度。仿真结果表明,所提方法具有很高的资源利用率。 The number of users in many open source IaaS platform,the system handle the tasks required quantity is very large,lead to the traditional particle swarm scheduling method,is easy to fall into local optimal solution to a certain range when a large number of redundant iteration,processing efficiency is low. Therefore,put forward a kind of based on the genetic and ant colony dynamic integration method of stochastic scheduling tasks in open source Iaa S platform,firstly analyzes the initialization pheromone of ant colony algorithm,and the choice of path,the update pheromone process in detail. Secondly introduces the chromosome coding genetic algorithm,the generation of initial population,selection operator,crossover operator and mutation operator of concrete steps. Then random task scheduling goal in open source Iaa S platform is given,solving out each of the open source Iaa S platform to complete the resources of the time required to complete the subtasks. Finally open source Iaa S platform task scheduling are formed by genetic algorithm is accurate initial solution,and with its initialized pheromone distribution,positive feedback and high efficiency implementation based on ant colony algorithm is convergent to obtain the advantages of the best random task scheduling. The simulation results show that the proposed method is of high resource utilization.
出处 《计算机仿真》 CSCD 北大核心 2015年第7期386-389,443,共5页 Computer Simulation
基金 院科研基金项目(2013KY18)
关键词 任务 调度 动态融合 Task Scheduling Dynamic fusion
  • 相关文献

参考文献10

二级参考文献105

共引文献76

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部