CC* Regional Computing: CENVAL-ARC: Central Valley Accessible Research and Computational Hub

CC* 区域计算：CENVAL-ARC：中央谷无障碍研究和计算中心

• 批准号: 2346744
• 负责人: Sarvani Chadalapaka
• 金额: $ 100万
• 依托单位: University of California - Merced
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2024
• 资助国家: 美国
• 起止时间: 2024-06-01 至 2026-05-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2346744&HistoricalAwards=false
• 关键词: CC; Regional; Computing; CENVAL; ARC

The Central Valley Advanced Research Computing (CENVAL-ARC) initiative aims to address the computational resource needs in the historically underserved Central Valley region of California, enhancing research and educational opportunities for a broad spectrum of students and researchers within Hispanic-Serving Central Valley institutions. The project is led by the University of California, Merced (UCM), in collaboration with California State Universities (CSUs) - CSU Sacramento, CSU Stanislaus, and CSU Fresno.CENVAL-ARC's hardware setup comprises Graphical Processing Unit (GPU) nodes and Central Processing Unit (CPU) nodes, each with distinct configurations of processor and memory specifications. The hardware seamlessly integrates into UCM's existing Pinnacles High Performance Computing (HPC) cluster. The cluster also features efficient fast scratch and large data storage, enhancing data storage and accessibility for science drivers. CENVAL-ARC utilizes Ethernet and Omnipath networks, supported by open-source tools like XDMoD, for efficient cluster management and data transfers. Its connection to a Science DMZ enhances data exchange with external institutions, fostering collaborative research. Running on RedHat Linux with the Slurm scheduler, CENVAL-ARC's computational resources accommodate diverse research tasks. Standard CPU nodes and large-memory CPU nodes support activities such as intricate systems modeling and numerical optimization, while GPUs are indispensable for executing double-precision floating-point calculations and managing Artificial Intelligence and Machine Learning (AI/ML) workloads.CENVAL-ARC plays a transformative role in advancing research throughout the Central Valley. The project supports yearly research symposia that engage dozens of regional participants and empower researchers across various scientific domains. This initiative aligns with the National Science Foundation's vision for catalyzing scientific transformation through cyberinfrastructure, fostering translational research and discovery.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.

中央谷高级研究计算 (CENVAL-ARC) 计划旨在解决历史上服务不足的加利福尼亚州中央谷地区的计算资源需求，为中央谷拉美裔服务机构内的广大学生和研究人员提供更多的研究和教育机会。 该项目由加州大学默塞德分校 (UCM) 牵头，与加州州立大学 (CSU) 合作 - CSU Sacramento、CSU Stanislaus 和 CSU Fresno。CENVAL-ARC 的硬件设置包括图形处理单元 (GPU) 节点和中央处理器处理单元 (CPU) 节点，每个节点都具有不同的处理器和内存规格配置。该硬件无缝集成到 UCM 现有的 Pinnacles 高性能计算 (HPC) 集群中。该集群还具有高效快速的暂存和大数据存储功能，增强了科学驾驶员的数据存储和可访问性。 CENVAL-ARC 利用以太网和 Omnipath 网络，并由 XDMoD 等开源工具支持，实现高效的集群管理和数据传输。它与科学 DMZ 的连接增强了与外部机构的数据交换，促进了合作研究。 CENVAL-ARC 的计算资源在带有 Slurm 调度程序的 RedHat Linux 上运行，可适应不同的研究任务。标准CPU节点和大内存CPU节点支持复杂系统建模和数值优化等活动，而GPU对于执行双精度浮点计算和管理人工智能和机器学习（AI/ML）工作负载是不可或缺的。CENVAL-ARC在推动整个中央山谷的研究方面发挥着变革性的作用。该项目支持每年举行的研究研讨会，吸引数十名地区参与者参与，并为各个科学领域的研究人员提供支持。这一举措符合美国国家科学基金会通过网络基础设施促进科学转型、促进转化研究和发现的愿景。该奖项反映了国家科学基金会的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。