MRI: Acquisition of High-Performance Computing Cluster for Research and Workforce Development at University of Cincinnati

MRI：收购辛辛那提大学用于研究和劳动力发展的高性能计算集群

• 批准号: 2018617
• 负责人: Prashant Khare
• 金额: $ 60万
• 依托单位: University of Cincinnati Main Campus
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-08-01 至 2023-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2018617&HistoricalAwards=false
• 关键词: MRI; Acquisition; Performance; Computing; Cluster

This project will the support the acquisition of a high-performance computing (HPC) and high-speed storage system at the University of Cincinnati (UC). This instrument will meet the computing needs of UC researchers and scientists spanning a broad range of fields including, physics, chemistry, aerospace & mechanical engineering, computer science, bioinformatics, medicine, and digital humanities. The research projects enabled by the HPC system in these areas will advance fundamental science and contribute to national priorities in health and national security. This system will also serve as a launching facility for the development of novel algorithms, and initial and scalability testing of codes, which will then be ported to national supercomputing facilities. Over 117 faculty and students will immediately benefit from this equipment. In addition to scientific advancements and development of the next generation of HPC-enabled workforce, this project will provide opportunities for the training of HPC administrators and facilitators. Through participation in the Open Science Grid, the equipment will also enable the research activities of external users. This project will involve the participation of underrepresented groups through summer internships, and workshops at local inner-city schools. Further, this project will offer the possibility of exposing researchers in humanities to the benefits of HPC through collaboration between the Advanced Research Computing and the Digital Scholarship Centers at UC.The computing cluster will consist of Central Processing Units (CPU), Graphics Processing Units (GPU) and high-speed scratch storage to substantially enhance the capabilities of the Advanced Research Computing Center at UC and thereby support computational and data science researchers by providing HPC resources for large-scale scientific calculations, HTC for Monte Carlo type simulations, and GPUs for advanced analytics, artificial intelligence, and visualization, all of which will enable sophisticated and increasingly realistic modeling, simulation and data analysis. These research efforts based on molecular dynamics and quantum chemistry simulations, computational fluid dynamics, multi-universe stochastic models, and data-intensive, deep learning and graph convolutional neural networks, will advance our knowledge of physical & cyber-physical systems, chemical and mechanical behaviors of materials, the structure and origin of our universe and contribute to the development of innovative applications such as brain-inspired computing, effective personalized precision medicine for complex diseases, carbon-neutral combustion devices, hypersonic propulsion systems, multi-UAV autonomous systems, and smart cities. These objectives closely align with one of UC’s enterprise-level programs, the Digital Futures initiative.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.

该项目将支持辛辛那提大学（UC）收购高性能计算（HPC）和高速存储系统。 该仪器将满足UC研究人员和科学家的计算需求，涵盖广泛的领域，包括物理，化学，航空航天机械工程，计算机科学，生物信息学，医学和数字人文。 HPC系统在这些领域实现的研究项目将推动基础科学的发展，并为国家在健康和国家安全方面的优先事项做出贡献。 这一系统还将作为开发新算法以及对代码进行初始测试和可扩展性测试的启动设施，然后将这些代码移植到国家超级计算设施。 超过117名教师和学生将立即受益于这些设备。 除了科学进步和下一代HPC支持的劳动力的发展外，该项目还将为HPC管理员和促进者提供培训机会。 通过参与开放科学网格，该设备还将使外部用户的研究活动成为可能。 这一项目将通过暑期实习和在当地市中心学校举办讲习班，让代表人数不足的群体参与。 此外，该项目将通过加州大学高级研究计算和数字奖学金中心之间的合作，为人文学科的研究人员提供接触HPC的可能性。计算集群将包括中央处理器（CPU），图形处理单元（GPU）和高性能加速暂存存储，以大幅增强加州大学高级研究计算中心的能力，从而支持计算和数据科学研究人员通过为大规模科学计算提供HPC资源，为Monte Carlo类型的模拟提供HTC，以及为高级分析、人工智能和可视化提供GPU，所有这些都将实现复杂且日益逼真的建模、模拟和数据分析。 这些研究工作基于分子动力学和量子化学模拟，计算流体动力学，多宇宙随机模型，以及数据密集型，深度学习和图形卷积神经网络，将推进我们对物理&网络物理系统，材料的化学和机械行为，我们宇宙的结构和起源，并有助于开发创新应用，如脑启发计算，针对复杂疾病的有效个性化精准医疗、碳中和燃烧装置、高超音速推进系统、多无人机自主系统和智慧城市。这些目标与加州大学的企业级计划之一--数字未来计划密切相关。该奖项反映了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。