Enzo: An Adaptive Mesh Refinement Code for Astrophysics
The Enzo Collaboration, Greg L. Bryan, Michael L. Norman, Brian W. O'Shea, Tom Abel, John H. Wise, Matthew J. Turk, Daniel R. Reynolds, David C. Collins, Peng W… 展开作者
The Enzo Collaboration, Greg L. Bryan, Michael L. Norman, Brian W. O'Shea, Tom Abel, John H. Wise, Matthew J. Turk, Daniel R. Reynolds, David C. Collins, Peng Wang, Samuel W. Skillman, Britton Smith, Robert P. Harkness, James Bordner, Ji-hoon Kim, Michael Kuhlen, Hao Xu, Nathan Goldbaum, Cameron Hummels, Alexei G. Kritsuk, Elizabeth Tasker, Stephen Skory, Christine M. Simpson, Oliver Hahn, Jeffrey S. Oishi, Geoffrey C So, Fen Zhao, Renyue Cen, Yuan Li
This paper describes the open-source code Enzo, which uses block-structured adaptive mesh refinement to provide high spatial and temporal resolution for modeling astrophysical fluid flows. The code is Cartesian, can be run in 1, 2, and 3 dimensions, and supports a wide variety of physics including hydrodynamics, ideal and non-ideal magnetohydrodynamics, N-body dynamics (and, more broadly, self-gravity of fluids and particles), primordial gas chemistry, optically-thin radiative cooling of primordial and metal-enriched plasmas (as well as some optically-thick cooling models), radiation transport, cosmological expansion, and models for star formation and feedback in a cosmological context. In addition to explaining the algorithms implemented, we present solutions for a wide range of test problems, demonstrate the code's parallel performance, and discuss the Enzo collaboration's code development methodology.
Trigonometric Extension of the Geometric Correction Factor: Prototype for adding precision to adaptive ray tracing in ENZO
在本文中,我们描述了一种方法,旨在增加精度的辐射模拟的自适应网格细化宇宙流体动力学代码ENZO。我们以\textit{ENZO+MORAY: radiation hydrodynamics adaptive mesh refinement simulations with adaptive ray tracing}(Wise and Abel 2011)中描述的几何校正因子为基础,该因子解释了立方体对射线立体角的部分覆盖。由于近似方法中的这种几何不匹配,辐射场中存在伪影。在这里,我们解决了二维扩展,它作为三维情况下的充分估计,在实践中,分层等面积等纬度像素化的球体(HEALPix)(Gorski 2005)。我们将证明几何校正因子的扩展的价值,并为将来实现ENZO以改进点源辐射的模拟奠定基础。
AMReX-Astrophysics Microphysics: A set of microphysics routines for astrophysical simulation codes based on the AMReX library
Collaboration(协作组);Stony Brook University(石溪大学);University of Wisconsin, Madison(威斯康星大学麦迪逊分校);Independent Researcher(独立研究者);University of Arizona(亚利桑那大学);The Australian National University(澳大利亚国立大学);Los Alamos National Laboratory(洛斯阿拉莫斯国家实验室);Lawrence Berkeley National Laboratory(劳伦斯伯克利国家实验室);Leiden Observatory(莱顿天文台);Arizona State University(亚利桑那州立大学);Michigan State University(密歇根州立大学);Institute for Advanced Computational Science, Stony Brook Univer(石溪大学高级计算科学研究所)
University of Amsterdam(阿姆斯特丹大学);KU Leuven(荷语鲁汶大学);Nanjing University(南京大学);Universidade de Aveiro(阿威罗大学);Centre for Research and Development in Mathematics and Applications (CIDMA)(数学与应用研究中心);Netherlands eScience Center(荷兰电子科学中心);Yunnan University(云南大学);Centrum Wiskunde & Informatica(数学与计算机科学中心)
Transferring Data from a Voronoi Mesh to an Adaptive Cartesian Grid in Pursuit of Self-consistent Top-down Star Formation
Drexel University(德雷塞尔大学);American Museum of Natural History(美国自然历史博物馆);Max Planck Institute for Astrophysics(马克斯·普朗克天体物理学研究所);McMaster University(麦克马斯特大学);Tsinghua University(清华大学);Columbia University(哥伦比亚大学);Leiden Observatory, Leiden University(莱顿大学莱顿天文台)
Parthenon -- a performance portable block-structured adaptive mesh refinement framework
在通往亿万级的道路上,计算机设备架构和相应的编程模型的前景变得更加多样化。虽然有各种低级别的性能可移植编程模型,但在应用程序级别上缺乏支持。为了解决这个问题,我们提出了性能便携式块结构自适应网格细化(AMR)框架帕台农神庙,来自经过测试和广泛使用的雅典娜++天体物理磁流体动力学代码,但广义作为基础的各种下游多物理场代码。Parthenon采用了Kokkos编程模型,并提供了从多维变量到定义和分离组件的包,再到启动并行计算内核的各种级别的抽象。Parthenon在设备内存中分配所有数据以减少数据移动,支持变量和网格块的逻辑打包以减少内核启动开销,并采用单侧异步MPI调用以减少多节点模拟中的通信开销。使用流体力学迷你应用程序,我们展示了各种架构上的弱和强扩展,包括AMD和NVIDIA GPU,Intel和AMD x86 CPU,IBM Power 9 CPU以及Fujitsu A64 FX CPU。在Frontier上的最大规模(第一台TOP500 exascale机器),miniapp在9,216个节点(73,728个逻辑GPU)上达到了$1.7\times10^{13}$区域周期/秒,弱扩展并行效率约为92%(从单个节点开始)。再加上它是一个开放的协作项目,这使得Parthenon成为一个理想的框架,以针对亿级模拟,其中下游开发人员可以专注于他们的特定应用,而不是处理并行,设备加速…