MRI: Acquisition of a Computer Cluster for the Study of Fluids, Heliophysics, and Astrophysical Plasmas

MRI：购买用于研究流体、太阳物理学和天体物理等离子体的计算机集群

• 批准号: 1919310
• 负责人: Francois Foucart
• 金额: $ 60.29万
• 依托单位: University of New Hampshire
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-09-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1919310&HistoricalAwards=false
• 关键词: MRI; Acquisition; Computer; Cluster; Study

A homogeneous high-performance computing cluster for the simulation of fluids, heliophysics, and astrophysical plasmas will be built at the University of New Hampshire (UNH). The fluid and plasma simulations performed at UNH share very similar hardware requirements, making it possible to design an instrument with a high expected usage rate and broad scientific applications. The cluster will play a critical role in the formation of UNH graduate students and postdoctoral fellows to modern computational methods and parallel computing, as well as for the recruitment and retention of a diverse workforce.Numerical simulations of fluids and plasmas play a critical role in the study of a wide range of problems in astrophysics, heliophysics, and geophysics. Modeling nearly-collisionless plasmas is necessary to understand the generation of the solar wind, its propagation, its interaction with the Earth's magnetosphere, and its effects on, for example, the Earth's climate or artificial satellites. The same physics governs the evolution of accretion disks around many of the supermassive black holes at the center of galaxies, including the soon-to-be imaged black hole at the center of our own Milky Way. Magnetohydrodynamics simulations, on the other hand, help us understand the evolution of the Earth's atmosphere and its oceans, as well as some of the highest-energy events observable in the Universe: the collisions of black holes and neutron stars that power gravitational wave signals. This grant will enable new understanding in all of these research areas through building a new computer cluster for numerical simulations of fluids and plasmas. The cluster will have 1760 compute nodes with Infiniband communication, and 283+ Tb of storage/archival space. This award was co-funded between the Division of Atmospheric and Geospace Sciences and the Division of Physics.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

将在新罕布什尔州大学（UNH）建立一个用于模拟流体、太阳物理学和天体物理学等离子体的同质高性能计算集群。在UNH进行的流体和等离子体模拟有着非常相似的硬件要求，这使得设计一种具有高预期使用率和广泛科学应用的仪器成为可能。该集群将发挥关键作用，在形成UNH的研究生和博士后研究员，以现代计算方法和并行计算，以及招聘和留住一个多元化的劳动力。流体和等离子体的数值模拟在天体物理学，太阳物理学和地球物理学的广泛问题的研究中发挥着关键作用。模拟几乎无碰撞的等离子体对于了解太阳风的产生、传播、与地球磁层的相互作用以及对地球气候或人造卫星等的影响是必要的。同样的物理学也控制着星系中心许多超大质量黑洞周围吸积盘的演化，包括我们银河系中心即将成像的黑洞。另一方面，磁流体动力学模拟有助于我们了解地球大气层和海洋的演变，以及宇宙中可观察到的一些最高能量事件：黑洞和中子星的碰撞，为引力波信号提供动力。这笔赠款将通过建立一个新的计算机集群，用于流体和等离子体的数值模拟，使所有这些研究领域的新的理解。该集群将拥有1760个具有Infiniband通信的计算节点，以及283+ Tb的存储/归档空间。 该奖项由大气和地球空间科学部和物理部共同资助，反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Global Simulations, chapter 37

全球模拟，第 37 章

• DOI: 10.1002/9781119815624
• 发表时间: 2021
• 期刊: Geography monograph series
• 作者: Joachim Raeder, Kai Germaschewski

Unsupervised classification of simulated magnetospheric regions

模拟磁层区域的无监督分类

• DOI: 10.5194/angeo-39-861-2021
• 发表时间: 2021
• 期刊: Annales Geophysicae
• 影响因子: 1.9
• 作者: Innocenti, Maria Elena;Amaya, Jorge;Raeder, Joachim;Dupuis, Romain;Ferdousi, Banafsheh;Lapenta, Giovanni
• 通讯作者: Lapenta, Giovanni

Domain of Influence Analysis: Implications for Data Assimilation in Space Weather Forecasting

• DOI: 10.3389/fspas.2020.571286
• 发表时间: 2020-09
• 作者: D. Millas;M. Innocenti;B. Laperre;J. Raeder;S. Poedts;G. Lapenta

Observational Analysis and Numerical Modeling of the Solar Wind Fluctuation Spectra during Intervals of Plasma Instability

等离子体不稳定期间太阳风脉动光谱的观测分析和数值模拟

• DOI: 10.3847/1538-4357/ac9f42
• 发表时间: 2022
• 期刊: The Astrophysical Journal
• 作者: Markovskii, S. A.;Vasquez, Bernard J.
• 通讯作者: Vasquez, Bernard J.

Inference and De-Noising of Non-Gaussian Particle Distribution Functions: A Generative Modeling Approach

• DOI: 10.1007/978-3-030-95467-3_25
• 发表时间: 2021-10
• 期刊: ArXiv
• 作者: J. Donaghy;K. Germaschewski