• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>2013 International Conference on Computer Communication and Informatics. >Meta-heuristic hybrid dynamic task scheduling in Heterogeneous computing environment
【24h】

Meta-heuristic hybrid dynamic task scheduling in Heterogeneous computing environment

机译：异构计算环境中的元启发式混合动态任务调度

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

System State estimation and decision making are the two major components of dynamic task scheduling in a distributed computing system. Heuristic and meta-heuristic approaches seem to be the most effective methods of scheduling in Heterogeneous computing due to their ability of relative fast generation of high quality solutions. Most of the available Meta heuristic algorithms attempt to find an optimal solution with respect to a specific fixed fitness measure. The major challenges when using Genetic Algorithms to solve dynamic optimization problems are: (a) to generate and keep the diversity in the populations, which is crucial for avoiding the premature convergence to the local optima and (b) to evolve robust solutions that are able to track the optima. All of these issues will necessitate the development of intelligent adaptive algorithms that can dynamically adapt to the changes in the large-scale Computing Groups. We propose a Hybrid Genetic and Case based reasoning algorithm HGAC to improve the make span by predicting the performance of online resources to better converging the local optima and improve decision faster in dynamic environment.
机译：系统状态估计和决策是分布式计算系统中动态任务调度的两个主要组成部分。启发式和元启发式方法似乎是异构计算中最有效的调度方法，因为它们能够快速生成高质量解决方案。大多数可用的元启发式算法都试图针对特定的固定适应性度量找到最佳解决方案。使用遗传算法解决动态优化问题时的主要挑战是：（a）生成并保持种群中的多样性，这对于避免过早收敛到局部最优是至关重要的；以及（b）发展能够跟踪最优值。所有这些问题将需要开发能够动态适应大型计算组变化的智能自适应算法。我们提出了一种基于混合遗传和案例的推理算法HGAC，通过预测在线资源的性能来改善生成范围，以更好地收敛局部最优值，并在动态环境中更快地改善决策。

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