MRI: Acquisition of a High-Performance Computing Cluster for Research and Teaching at UC Irvine

此次收购提供了一个现代化的、节能的、全校范围的高性能计算社区集群（HPC3），使从基本粒子物理和量子化学到地球系统建模和工程再到社会生态学等领域的研究和研究培训成为可能。HPC3有望为大规模并行和共享内存应用程序提供高灵活性和用户友好性以及高效率的组合，而集中式超级计算或云计算设施并不总是能够实现这些应用程序。此次收购是一个混合CPU/GPU集群，为13个主要用户研究小组以及该机构和加州州立大学三个当地合作校区的其他研究人员和学生组成的多样化高性能计算（HPC）社区提供强大的计算能力。通过HPC3与开放科学网格（OSG）和极端科学与工程发现环境（XSEDE）的集成，此次收购还将被全国各地的研究人员使用。将使用HPC3的研究项目示例包括：新基本粒子的发现、高温超导体和量子自旋液体的第一性原理模拟、热致密物质的模拟、小间隙化合物的电子结构、光催化剂和光驱动材料的非绝热分子动力学模拟、拥挤蛋白质溶液的蒙特卡罗模拟、药物的热力学性质、RNA测序、生物网络的排列、冰盖模型、全球气候模拟，包括云效应、地球物理气候数据分析、多尺度材料建模和湍流模拟。所有合作大学都是为西班牙裔学生服务的机构，在科学、技术、工程和数学（STEM）领域代表性不足的少数族裔学生将通过在校园和服务不足地区的当地高中进行指导和推广来积极招募。HPC3还将支持高度优化的HPC软件的开发，供全球科学家用于凝聚态物理、量子化学、计算流体动力学和地球系统科学的模拟。该奖项反映了美国国家科学基金会的法定使命，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。

This acquisition provides a modern, energy-efficient, campus-wide, High-Performance Computing Community Cluster (HPC3) enabling research and research training in areas ranging from elementary particle physics and quantum chemistry through earth-system modeling and engineering to social ecology. HPC3 is expected to deliver a combination of high flexibility and user friendliness with high efficiency for both massively parallel and shared memory applications that is not always achievable by centralized supercomputing or cloud computing facilities.The acquisition is a mixed CPU/GPU cluster that provides substantial computing power to a diverse, high-performance computing (HPC) community of 13 major user research groups as well as other researchers and students at the institution and three local partnering campuses of California State University. The acquisition will also be used by researchers across the Nation by virtue of HPC3's integration with the Open Science Grid (OSG) and eXtreme Science and Engineering Discovery Environment (XSEDE). Examples of research projects that will use HPC3 include: new elementary particle discovery, first-principles simulations of high-temperature superconductors and quantum spin liquids, simulations of warm dense matter, electronic structure of small-gap compounds, nonadiabatic molecular dynamics simulations of photocatalysts and light-driven materials, Monte Carlosimulations of crowded protein solutions, thermodynamic properties of pharmaceuticals, RNA sequencing, alignment of biological networks, ice-sheet modeling, global climate modeling including cloud effects, geophysical climate data analysis, multiscale materials modeling, and simulations of turbulent flow. All partnering universities are Hispanic-serving institutions, and minorities underrepresented in the Science, Technology, Engineering, and Mathematics (STEM) fields will be actively recruited through mentoring and outreach on campus and to local high schools in underserved areas. HPC3 will also support the development of highly optimized HPC software used by scientists worldwide for simulations in condensed matter physics, quantum chemistry, computational fluid dynamics, and earth system science.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.