首页> 外文OA文献 >Exploiting multi-scale parallelism for large scale numerical modelling of laser wakefield accelerators
【2h】

Exploiting multi-scale parallelism for large scale numerical modelling of laser wakefield accelerators

机译:利用多尺度并行机制进行大规模数值模拟   激光尾场加速器

代理获取
本网站仅为用户提供外文OA文献查询和代理获取服务,本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文,但由于OA文献来源多样且变更频繁,仍可能出现获取不到、文献不完整或与标题不符等情况,如果获取不到我们将提供退款服务。请知悉。

摘要

A new generation of laser wakefield accelerators, supported by the extremeaccelerating fields generated in the interaction of PW-Class lasers andunderdense targets, promises the production of high quality electron beams inshort distances for multiple applications. Achieving this goal will relyheavily on numerical modeling for further understanding of the underlyingphysics and identification of optimal regimes, but large scale modeling ofthese scenarios is computationally heavy and requires efficient use ofstate-of-the-art Petascale supercomputing systems. We discuss the maindifficulties involved in running these simulations and the new developmentsimplemented in the OSIRIS framework to address these issues, ranging frommulti-dimensional dynamic load balancing and hybrid distributed / shared memoryparallelism to the vectorization of the PIC algorithm. We present the resultsof the OASCR Joule Metric program on the issue of large scale modeling of LWFA,demonstrating speedups of over 1 order of magnitude on the same hardware.Finally, scalability to over $\sim 10^6$ cores, and sustained performance over$\sim 2$ PFlops is demonstrated, opening the way for large scale modeling oflaser wakefield accelerator scenarios.
机译:新一代激光尾波场加速器,在PW级激光器和低密度靶相互作用中产生的极强加速场的支持下,有望在短距离内生产出高质量的电子束,以用于多种应用。实现此目标将极大地依赖于数值模型来进一步理解基础物理和确定最佳方案,但是这些方案的大规模建模在计算上很繁琐,需要有效使用最新的Petascale超级计算系统。我们讨论了运行这些仿真所涉及的主要困难以及OSIRIS框架中为解决这些问题而实现的新开发,这些开发从多维动态负载平衡和混合分布式/共享内存并行性到PIC算法的矢量化。我们展示了OASCR Joule Metric程序在LWFA大规模建模问题上的结果,证明了在同一硬件上的加速超过1个数量级。最后,可扩展性超过$ \ sim 10 ^ 6 $内核,并且性能持续超过演示了$ \ sim 2 $ PFlops,为大规模建模激光尾场加速器场景开辟了道路。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利

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