CC* Compute: A High Performance GPU Cluster at Syracuse University

CC* 计算：雪城大学的高性能 GPU 集群

• 批准号: 2018822
• 负责人: Samuel Scozzafava
• 金额: $ 39.31万
• 依托单位: Syracuse 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: 2020-09-01 至 2022-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2018822&HistoricalAwards=false
• 关键词: CC; Compute; Performance; GPU; Cluster

Syracuse University is constructing a new compute cluster using Graphics Processing Units. This production cluster will serve the needs of the Syracuse University research community and the broader scientific community through integration with the Open Science Grid. The use of Graphics Processing Units to solve problems in science and engineering has grown significantly over recent years. Campus cyberinfrastructure providers have seen increasing demand for access to Graphics Processing Units in lieu of more traditional compute architectures. Syracuse University has created a strong partnership among campus-level cyberinfrastructure experts as well as at the national level through the University's contributions to the Open Science Grid. This cluster provides a significant resource for enabling science and cyberinfrastructure development at Syracuse University. The application drivers come from a variety of domains including computational forensics, high-energy physics, development of smart vision systems, computational chemistry, biomedical engineering, soft-matter physics, and gravitational-wave physics.Access to the Graphics Processing Units will be an important resource for education of undergraduate and graduate students. Given the growth in related Graphics Processing Unit computing, Syracuse University faculty are introducing Graphics Processing Unit programming to undergraduate students from a variety of backgrounds. The capabilities and availability of this cluster will allow for broader adoption within the classroom environment, as well as provide compute power to researchers. Undergraduates from across the College of Engineering and Computer Science and the College of Arts and Sciences can use these resources, giving them early preparation for new computing schemes. Access to the resources via the Open Science Grid will allow researchers from across the U.S. access to the compute resources.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.

锡拉丘兹大学正在使用图形处理单元构建一个新的计算集群。这个生产集群将通过与开放科学网格的集成来满足锡拉丘兹大学研究界和更广泛的科学界的需求。近年来，使用图形处理单元来解决科学和工程问题的数量显著增加。校园网络基础设施提供商已经看到越来越多的需求访问图形处理单元，而不是更传统的计算架构。锡拉丘兹大学通过对开放科学网格的贡献，在校园级网络基础设施专家之间以及国家一级建立了强有力的伙伴关系。该集群为锡拉丘兹大学的科学和网络基础设施发展提供了重要资源。应用驱动程序来自各种领域，包括计算取证，高能物理，智能视觉系统的开发，计算化学，生物医学工程，软物质物理和引力波物理。访问图形处理单元将是本科生和研究生教育的重要资源。鉴于相关图形处理单元计算的增长，锡拉丘兹大学的教师正在向来自各种背景的本科生介绍图形处理单元编程。该集群的功能和可用性将允许在课堂环境中更广泛地采用，并为研究人员提供计算能力。来自工程与计算机科学学院和艺术与科学学院的本科生可以使用这些资源，为他们早期准备新的计算方案。通过开放科学网格访问资源将允许来自美国各地的研究人员访问计算资源。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。