首页> 外文会议>2010 IASTED technology conferences >IMPROVED PARALLEL GENETIC ALGORITHM FOR TASK SCHEDULING ON HETEROGENEOUS PARALLEL SYSTEM
【24h】

IMPROVED PARALLEL GENETIC ALGORITHM FOR TASK SCHEDULING ON HETEROGENEOUS PARALLEL SYSTEM

机译:异类并行系统任务调度的改进并行遗传算法。

获取原文
获取原文并翻译 | 示例

摘要

The scheduling problem deals with the optimalrnassignment of a set of tasks onto parallel system andrnorders their execution so that the total completion time isrnminimized. It is well known NP-complete problem. Thisrnpaper proposes a genetic based approach to schedulernparallel tasks on heterogeneous parallel multiprocessorrnsystem. The scheduling problem considered in this paperrnincludes - next to search for an optimal mapping of therntask and their sequence of execution and also search forrnan optimal configuration of the parallel system. Anrnapproach for the simultaneous optimization of all thesernthree components of scheduling method using improvedrnparallel genetic algorithm is presented and itsrnperformance is evaluated in comparison with traditionalrnscheduling methods.
机译:调度问题涉及一组任务在并行系统上的最优分配,并对其执行进行排序,从而使总完成时间最小化。这是众所周知的NP完全问题。本文提出了一种基于遗传的方法来调度异构并行多处理器系统上的并行任务。本文考虑的调度问题包括-接下来是搜索任务的最佳映射及其执行顺序,还包括搜索并行系统的最佳配置。提出了一种使用改进的并行遗传算法同时优化调度方法的所有三个组成部分的方法,并与传统调度方法进行了比较。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号