MRI: Acquisition of a High-Performance Computational System for OAK Region to Enable Computing and Data Driven Discovery

MRI：为 OAK 地区采购高性能计算系统，以实现计算和数据驱动的发现

• 批准号: 2216084
• 负责人: Pratul Agarwal
• 金额: $ 400万
• 依托单位: Oklahoma State University
• 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: 2022-08-15 至 2025-07-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2216084&HistoricalAwards=false
• 关键词: MRI; Acquisition; Performance; Computational; System

This project will acquire and deploy a high-performance computing (HPC) system to serve the researchers and educators in the broad Oklahoma-Arkansas-Kansas (OAK) region. Computational modeling, simulations, and data analytics are essential tools for new discoveries across all areas of science, engineering and mathematics. In particular, modern research breakthroughs are being driven by vast amounts of data enabled by computing and memory capacity, often combined with machine-learning/artificial intelligence (ML/AI) techniques. This system will seed the growth of a collaborative HPC ecosystem in the region, serving the needs of experienced users and enabling first time users, especially from small and under-served institutions. Technical know-how exchange, regular seminars, and conferences will promote collaborative research endeavors of regional and national importance. Hands on training of graduate, undergraduate and high-school students in computational and data sciences will broadly improve the skill-sets of the science and engineering workforce in the OAK region. The deployed system will harness the power of the latest CPUs, cutting-edge graphics processing units (GPUs), 100 Terabytes of aggregate memory, an HDR InfiniBand interconnect, and Petabyte-scale high-speed storage. This system consisting of several types of nodes (CPUs-only, mid-range GPUs, and high-end GPUs) will provide close to 100 million-core hours of computing and will serve as a vital regional resource for science and engineering research. Initial research areas to be served by the system include biology, human and animal health, agriculture, environment research, chemistry and chemical engineering, semiconductor materials research, cybersecurity and social network modeling, renewable energy research, seismology, high-energy physics, and medical physics. This significantly improved computing and data analysis capabilities provided by this system will greatly benefit scientific productivity. Furthermore, the instrument will make it possible for faculty to include HPC and data analysis techniques in the STEM education curriculum, leading to computational thinking becoming a part of research in the minds of the next generation of scientists. Lowering the barrier of entry to HPC will also help improve the education and training of women, minorities and under-represented groups. This system will also contribute computing resources beyond the OAK region through collaborative mechanisms such as the Partnership to Advance Throughput computing (PATh) and the Open Science Grid (OSG).This project is jointly funded by the Major Research Instrumentation (MRI) program, the Established Program to Stimulate Competitive Research (EPSCoR), and the Computer & Information Science & Engineering (CISE) Directorate.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.

该项目将获得并部署一个高性能计算（HPC）系统，为俄克拉荷马州-阿肯色州-堪萨斯（OAK）地区的研究人员和教育工作者提供服务。计算建模、模拟和数据分析是科学、工程和数学各个领域新发现的重要工具。特别是，现代研究突破是由计算和内存容量支持的大量数据驱动的，通常与机器学习/人工智能（ML/AI）技术相结合。该系统将为该地区的协作HPC生态系统的发展奠定基础，满足经验丰富的用户的需求，并使首次用户，特别是来自小型和服务不足的机构的用户能够使用。技术知识交流，定期研讨会和会议将促进区域和国家重要性的合作研究工作。对研究生、本科生和高中生进行计算和数据科学方面的实践培训，将广泛提高OAK地区科学和工程劳动力的技能。 部署的系统将利用最新的CPU，尖端的图形处理单元（GPU），100 TB的聚合内存，HDR InfiniBand互连和PB级高速存储的功能。该系统由几种类型的节点（仅CPU、中档GPU和高端GPU）组成，将提供近1亿核心小时的计算，并将成为科学和工程研究的重要区域资源。该系统最初服务的研究领域包括生物学、人类和动物健康、农业、环境研究、化学和化学工程、半导体材料研究、网络安全和社会网络建模、可再生能源研究、地震学、高能物理和医学物理。该系统提供的计算和数据分析能力的显著提高将极大地有利于科学生产力。此外，该仪器将使教师能够将HPC和数据分析技术纳入STEM教育课程，从而使计算思维成为下一代科学家头脑中研究的一部分。降低进入HPC的门槛也将有助于改善妇女、少数民族和代表性不足群体的教育和培训。该系统还将通过合作机制，如合作伙伴关系，以促进物联网计算（PATh）和开放科学网格（OSG）的OAK地区以外的计算资源。该项目是由重大研究仪器（MRI）计划，建立计划，以刺激竞争性研究（EPSCoR）和计算机&信息科学&工程（CISE）联合资助。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。