【24h】

A Performance Model of Non-deterministic Particle Transport on Large-Scale Systems

机译:大型系统上不确定粒子传输的性能模型

获取原文
获取原文并翻译 | 示例

摘要

In this work we present a predictive analytical model that encompasses the performance and scaling characteristics of a non-deterministic particle transport application, MCNP. Previous studies on the scalability of parallel Monte Carlo eigenvalue calculations have been rather general in nature. It can be used for the simulation of neutron, photon, electron, or coupled transport, and has found uses in many problem areas. The performance model is validated against measurements on an AlphaServer ES40 system showing high accuracy across many processor /problem combinations. It is parametric with both application characteristics (e.g. problem size), and system characteristics (e.g. communication latency, bandwidth, achieved processing rate) serving as input. The model is used to provide insight into the achievable performance that should be possible on systems containing thousands of processors and to quantify the impact that possible improvements in sub-system performance may have. In addition, the impact on performance of modifying the communication structure of the code is also quantified.
机译:在这项工作中,我们提出了一个预测性分析模型,其中包含非确定性粒子传输应用程序MCNP的性能和缩放特性。本质上,以前关于并行蒙特卡洛特征值计算的可伸缩性的研究相当普遍。它可以用于模拟中子,光子,电子或耦合传输,并已在许多问题领域中得到了应用。该性能模型已针对AlphaServer ES40系统上的测量进行了验证,表明在许多处理器/问题组合中均具有很高的准确性。它以应用程序特征(例如问题大小)和系统特征(例如通信等待时间,带宽,达到的处理速率)作为输入参数化。该模型用于深入了解在包含数千个处理器的系统上应可实现的性能,并量化子系统性能可能提高可能带来的影响。另外,还可以量化修改代码的通信结构对性能的影响。

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