首页> 外文期刊>Parallel and Distributed Computing Practices >PARALLEL ADAPTIVE MESH REFINEMENT WITH LOAD BALANCING ON HETEROGENEOUS CLUSTER
【24h】

PARALLEL ADAPTIVE MESH REFINEMENT WITH LOAD BALANCING ON HETEROGENEOUS CLUSTER

机译:异构集群上具有负载平衡的并行自适应网格细化

获取原文
获取原文并翻译 | 示例
           

摘要

The efficient solution of many large-scale scientific calculations depends on unstructured mesh strategies. For example, problems where the solution changes rapidly in small regions of the domain require an adaptive mesh strategy. In this paper we discuss the main algorithmic issues to be addressed with an integrated approach to solving these problems on heterogeneous parallel architectures. We review new parallel algorithms to solve two significant problems that arise in this context: the mesh refinement, partitioning and load balancing. The procedure to support parallel refinement and redistribution of two dimensional unstructured finite element meshes on heterogeneous computers is presented. The solver is based on the parallel conjugate gradient method. The error indicator and the resulting refinement parameters are computed in parallel.
机译:许多大规模科学计算的有效解决方案取决于非结构化网格策略。例如,解决方案在域的小区域中快速变化的问题需要自适应网格策略。在本文中,我们讨论了使用综合方法来解决异构并行体系结构上的这些问题的主要算法问题。我们回顾了新的并行算法,以解决在此情况下出现的两个重要问题:网格细化,分区和负载平衡。提出了在异构计算机上支持二维非结构化有限元网格的并行细化和重新分布的过程。求解器基于并行共轭梯度法。错误指示器和最终的优化参数是并行计算的。

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