MRI: Acquisition of a high-performance computing resource to enhance research and undergraduate education at the College of Staten Island

MRI：收购高性能计算资源以加强史坦顿岛学院的研究和本科教育

• 批准号: 2215760
• 负责人: Angelo Bongiorno
• 金额: $ 72.35万
• 依托单位: CUNY College of Staten Island
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2215760&HistoricalAwards=false
• 关键词: MRI; Acquisition; performance; computing; resource

This award supports the acquisition of a high-performance computing (HPC) cluster to enhance research and undergraduate education at the College of Staten Island (City University of New York). The HPC cluster will serve a diverse pool of faculty to boost computational research in areas ranging from biology and biomedicine to computer science, engineering, and materials science. By aiding studies characterizing large biomolecules, this resource will help garner an understanding of basic life processes relevant to human health and the environment. Likewise, researchers working in computer science, engineering, and materials science will benefit in their development of new algorithms for the analysis and extraction of information from complex and large digital datasets. Coupled with the application of modern simulation techniques to investigate minerals of geophysical relevance, such endeavors will lead to a deeper insight into the composition of the Earth's mantle. The acquisition will also support and expand the educational activities leveraging high-performance computers as tools for learning and training, including coursework in computational and STEM disciplines by offering hands-on computing activities to high school students and teachers.The computational instrument will consist of two computing ‘super’ nodes with a hybrid architecture comprising of both central and graphics processing units (CPUs and GPUs), integrated with a large-capacity intelligent storage system. This HPC resource is tailored to support a range of data-heavy parallel workflows, which will enable investigators to augment scientific knowledge in areas where experimental science is reaching its limits, and where progress is intrinsically driven by computational studies. This cluster will enable: (i) novel computationally-enhanced cryo-electron microscopy techniques to unravel the structure and dynamics of biomolecules; (ii) modern atomistic simulations to compute binding affinities of chemical compounds to protein receptors and hence discover new drug therapies; (iii) development of new data mining and artificial intelligence (AI) algorithms for the analysis and predictive modeling of complex networks, high-dimensional data and 3D objects; and (iv) exploit quantum mechanical calculations to predict the properties of rock-forming minerals of geophysical relevance. The HPC cluster will also enable educators to augment formal classroom education in computational disciplines, stimulate undergraduate curriculum development, and support graduate and undergraduate research projects. Access to this resource will prepare students and future generation of scientists for diverse careers in research and industry that are increasingly relying on the use of computers, cloud technologies, and parallel computing.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.

该奖项支持收购高性能计算（HPC）集群，以加强史坦顿岛学院（纽约城市大学）的研究和本科教育。高性能计算集群将为多样化的教师提供服务，以促进从生物学和生物医学到计算机科学、工程和材料科学等领域的计算研究。通过协助表征大分子的研究，这一资源将有助于了解与人类健康和环境有关的基本生命过程。同样，在计算机科学、工程和材料科学领域工作的研究人员将受益于他们从复杂和大型数字数据集中分析和提取信息的新算法的开发。再加上应用现代模拟技术来研究与地球物理相关的矿物，这样的努力将导致对地幔组成的更深入的了解。此次收购还将支持和扩大利用高性能计算机作为学习和培训工具的教育活动，包括通过为高中学生和教师提供动手计算活动来完成计算和STEM学科的课程。该计算仪器将由两个计算“超级”节点组成，其混合架构包括中央和图形处理单元（cpu和gpu），并集成了一个大容量智能存储系统。这种高性能计算资源是为支持大量数据并行工作流程而量身定制的，这将使研究人员能够在实验科学达到极限的领域增加科学知识，并且在本质上由计算研究驱动的领域取得进展。该集群将使：(i)新的计算增强低温电子显微镜技术来揭示生物分子的结构和动力学；（ii）现代原子模拟，以计算化合物与蛋白质受体的结合亲和力，从而发现新的药物疗法；（iii）开发新的数据挖掘和人工智能（AI）算法，用于复杂网络、高维数据和3D对象的分析和预测建模；（iv）利用量子力学计算来预测与地球物理相关的造岩矿物的性质。HPC集群还将使教育者能够增加计算学科的正规课堂教育，刺激本科课程的开发，并支持研究生和本科生的研究项目。获取这一资源将为学生和未来一代的科学家在越来越依赖于计算机、云技术和并行计算的研究和工业领域的不同职业生涯做好准备。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。