CC* Compute: A high-performance computing cluster to accelerate research, education, and training at Rhodes College

罗德学院正在建设和部署一个高性能计算集群，作为一个校园范围内的研究计算中心的基石。该项目建立了一个必要的计算资源，为计算机科学、生物学、数学、化学和经济学领域的研究人员和学生提供了新的能力。该集群提供了一种新的计算资源，其规模在以前的校园中是不可用的，并提高了机构支持研究的能力，包括分析线粒体DNA，模拟无机化合物，以及提高并行和分布式系统的性能和可用性。作为一所研究密集型文理学院，罗德学院的这个集群系统极大地促进了学生与广泛学科的研究人员一起从事基于计算的科学研究的机会。这个集群系统还可以作为课堂体验的平台，帮助学生理解如何使用计算方法来更好地理解世界。该系统支持计算机科学、生物学和数学的课堂活动。此外，该项目使研究人员能够通过XSEDE在国家计算资源上准备他们的实验，并将未使用的计算资源贡献给开放科学网格，这是一个国家分布式计算合作伙伴关系。集群系统支持高性能和高吞吐量的工作负载，系统总容量为1,584核、9TB内存和504tb存储。集群由1台登录服务器、1台存储服务器、44个计算节点和100Gbit/s ib互连网络组成。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。

Rhodes College is building and deploying a high-performance computing cluster that serves as the cornerstone of a campus-wide research computing center. This project establishes a needed computing resource that enables new capabilities for researchers and students in computer science, biology, mathematics, chemistry, and economics. The cluster provides a new computing resource at a scale not previously available on campus and advances the institutions' ability to support research, including analyzing mitochondrial DNA, modeling inorganic compounds, and improving the performance and usability of parallel and distributed systems. As a research-intensive liberal arts college, this cluster system at Rhodes College dramatically advances the opportunities for students to be engaged in computation-based scientific research alongside researchers in a wide range of disciplines. This cluster system also functions as a platform for classroom experiences that help students understand how computational methods can be used to better understand the world. The system supports class activities in computer science, biology, and mathematics. Additionally, this project enables researchers to prepare their experiments to run on national compute resources through XSEDE and will also contribute unused computing resources to the Open Science Grid, a national, distributed computing partnership.The cluster system supports both high-performance and high-throughput workloads with an overall system capacity of 1,584 cores, 9TB of memory, and 504 TB of storage. The cluster is comprised of a login server, storage server, 44 compute nodes, and a 100Gbit/s InfiniBand interconnect network.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.