CC* Compute: GPU-based Computation and Data Enabled Research and Education (G-CoDERE) at PSU

CC* 计算：PSU 基于 GPU 的计算和数据支持的研究和教育 (G-CoDERE)

• 批准号: 2019216
• 负责人: Feng Liu
• 金额: $ 39.59万
• 依托单位: Portland State University
• 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: 2020-07-01 至 2022-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2019216&HistoricalAwards=false
• 关键词: CC; Compute; GPU; based; Computation

Data and computationally intensive research and education is increasingly important at Portland State University (PSU). Scientists and students at PSU are producing massive quantities of data and investigating machine learning and data science approaches to research problems in many different fields. Graphics processing units (GPUs) excel at large-scale parallel computing and are critical for analyzing and visualizing such massive quantities of research data and developing data-driven technologies. This project establishes PSU's first GPU computing infrastructure and will support PSU research groups in a wide variety of fields, including Computer Science, ECE, Physics, Chemistry, Statistics, and Speech & Hearing Science, and benefit researchers in partner universities, including Oregon Health and Science University and Lewis & Clark University. It provides undergraduates, graduate students, and postdoctoral researchers new training opportunities for data-driven research; enables the creation of new machine learning, data analytics, and visualization courses; and supports upgrading existing courses with emerging data-driven paradigm. It facilitates the K12 outreach programs such as Oregon Mathematics, Engineering, Science Achievement and Saturday Academy’s Apprenticeships in Science and Engineering with GPU-enabled project and internship opportunities. This infrastructure allows PSU, Oregon's most diverse public university, to provide the state-of-the-art GPU facility and learning opportunities to students from underrepresented groups.This project establishes PSU's first GPU computing infrastructure by acquiring twenty GPU servers with the related high-performance data storage. This GPU infrastructure complements PSU's Coeus high-performance computing cluster to support GPU-enabled research and education at PSU and share with external users through the Open Science Grid.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.

数据和计算密集型的研究和教育在波特兰州立大学（PSU）越来越重要。PSU的科学家和学生正在产生大量数据，并研究机器学习和数据科学方法，以研究许多不同领域的问题。图形处理器（GPU）擅长大规模并行计算，对于分析和可视化如此大量的研究数据以及开发数据驱动技术至关重要。该项目建立了PSU的第一个GPU计算基础设施，并将支持PSU在各种领域的研究小组，包括计算机科学，ECE，物理，化学，统计和语音听力科学，并使合作大学的研究人员受益，包括俄勒冈州健康与科学大学和刘易斯克拉克大学。它为本科生，研究生和博士后研究人员提供了数据驱动研究的新培训机会;能够创建新的机器学习，数据分析和可视化课程;并支持用新兴的数据驱动范式升级现有课程。它促进了K12外展计划，如俄勒冈州数学，工程，科学成就和周六学院的科学和工程学徒与GPU启用的项目和实习机会。该基础设施使俄勒冈州最多元化的公立大学PSU能够为代表性不足的群体的学生提供最先进的GPU设施和学习机会。该项目通过购买二十台GPU建立了PSU的第一个GPU计算基础设施服务器以及相关的高性能数据存储。该GPU基础设施补充了PSU的Coeus高性能计算集群，以支持PSU支持GPU的研究和教育，并通过开放科学网格与外部用户共享。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。