• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Conference on physics of reactors >RESEARCH AND IMPLEMENTATION OF HYBRID PARALLEL ALGORITHM IN MONTE CARLO CRITICALITY AND BURNUP CALCULATION
【24h】

RESEARCH AND IMPLEMENTATION OF HYBRID PARALLEL ALGORITHM IN MONTE CARLO CRITICALITY AND BURNUP CALCULATION

机译：蒙特卡罗临界度和燃耗计算的混合并行算法研究与实现。

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

With the continuous development of computer hardware, high-performance Monte Carlo transport solver has become a focus recently. A hybrid message-passing and shared-memory approach, which enables the solver to utilize available computing resources both between and within compute nodes on supercomputer, has become the conclusive parallel architecture for MC solver. Shared-memory parallel model can be well suitable for multi -core system in the future. This paper is to summarize the recent advancements of Reactor Monte Carlo code RMC in parallel performance study, including the message passing parallel model, shared-memory parallel model and hybrid parallel based on domain replication strategy. The parallel efficiency and scalability of the models is established through a series of scaling studies, performed with RMC, which cover a variety of representative scenarios for large-scale reactor criticality and depletion simulations. Results from these studies demonstrate that RMC, using the hybrid parallel model described in this paper, is able to handle pin-wise full core burnup calculations with millions of depletion regions and scales well through thousands of processors on MilkyWay-2 supercomputer, which is ranked NO.1 among World's TOP500 Supercomputers.
机译：随着计算机硬件的不断发展，高性能的蒙特卡洛运输求解器已成为近期的研究重点。混合消息传递和共享内存方法使求解器能够利用超级计算机上计算节点之间和内部的可用计算资源，已成为MC求解器的最终并行架构。共享内存并行模型将来可能非常适合多核系统。本文旨在总结Reactor蒙特卡罗代码RMC在并行性能研究中的最新进展，包括基于域复制策略的消息传递并行模型，共享内存并行模型和混合并行。该模型的并行效率和可扩展性是通过使用RMC进行的一系列缩放研究建立的，该研究涵盖了大规模反应堆关键性和耗竭模拟的各种代表性方案。这些研究的结果表明，使用本文所述的混合并行模型，RMC能够处理具有数百万个耗尽区的逐针全核燃耗计算，并能够通过MilkyWay-2超级计算机上的数千个处理器进行很好的扩展，该排名在世界TOP500超级计算机中排名第一。

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