Collaborative Research: IRNC: Testbed: FAB: FABRIC Across Borders

合作研究：IRNC：测试平台：FAB：FABRIC 跨境

• 批准号: 2029261
• 负责人: Paul Ruth
• 金额: $ 177.54万
• 依托单位: University of North Carolina at Chapel Hill
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-09-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2029261&HistoricalAwards=false
• 关键词: Collaborative; Research; IRNC; Testbed; FAB

Global science relies on robust, interconnected components - computers, storage, networks and the software that ties them together - collectively called the scientific cyberinfrastructure (CI). Improvements to individual components are made at varying paces, often creating bottlenecks in the flow of information - the scientific workflow - and slowing down scientific discovery. FABRIC Across Borders (FAB) enables domain scientists and CI experts to jointly develop a more tightly integrated, flexible, intelligent, easily programmable workflow that takes advantage of rapid changes in technology to improve global science collaboration. FAB enables domain scientists to perform global, end-to-end experimentation of new CI workflow ideas on a platform with one of a kind capabilities. The project expands the NSF-funded FABRIC testbed to encompass four additional, International locations, creating an interconnected resource on which an initial set of scientists from High Energy Physics (HEP), Astronomy, Cosmology, Weather, Urban Science and Computer Science work with cyberinfrastructure experts to conduct cyberinfrastructure experiments. In addition to domain scientists, FAB collaborates in the area of Internet freedom and maintains strong partnerships with human rights groups, which serve to expand the results beyond domain sciences.FABRIC nodes contain programmable networking hardware, storage, CPUs and GPUs, measurement devices and software in a single, integrated rack. FAB enables placement of four additional nodes in partner data centers in Tokyo, Amsterdam, Bristol and the particle physics lab CERN in Geneva and connects them via NSF-funded International networks, on which it’s possible to conduct experiments without impacting production science. FAB offers programmable peering with production networks and specialized testbeds, allowing experimenter topologies to be joined with production networks, vastly expanding the possibilities for the types of resources and users that can utilize the infrastructure. FAB creates new software services and tools for researchers at the facilities, and interfaces with existing and evolving data delivery services to efficiently move and process scientific data globally and test novel data analysis approaches that scale to massive volumes. Metrics of success are driven by the science experiments themselves: more efficient handling of both high energy physics data from CERN experiments to worldwide collaborators and Cosmic Microwave Background data collected in South America and the South Pole; successful proofs of concept for the sharing of Smart City sensor data for urban planning as well as the establishment of global, private 5G networks. All software associated with FAB will be open source and posted in a publicly available repository: https://github.com/fabric-testbed/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.

全球科学依赖于强大的、互连的组件——计算机、存储、网络以及将它们连接在一起的软件——统称为科学网络基础设施（CI）。对各个组件的改进速度各不相同，通常会在信息流（科学工作流程）中造成瓶颈，并减慢科学发现的速度。 FABRIC Across Borders (FAB) 使领域科学家和 CI 专家能够共同开发更紧密集成、灵活、智能、易于编程的工作流程，利用技术的快速变化来改善全球科学协作。 FAB 使领域科学家能够在具有一种功能的平台上对新的 CI 工作流程想法进行全局、端到端的实验。该项目将 NSF 资助的 FABRIC 测试平台扩展至另外四个国际地点，创建了一个互连的资源，来自高能物理 (HEP)、天文学、宇宙学、天气、城市科学和计算机科学的首批科学家与网络基础设施专家一起开展网络基础设施实验。除了领域科学家之外，FAB 还在互联网自由领域开展合作，并与人权组织保持牢固的伙伴关系，这有助于将成果扩展到领域科学之外。FABRIC 节点在单个集成机架中包含可编程网络硬件、存储、CPU 和 GPU、测量设备和软件。 FAB 能够在东京、阿姆斯特丹、布里斯托尔的合作伙伴数据中心以及日内瓦的粒子物理实验室 CERN 中放置四个额外的节点，并通过 NSF 资助的国际网络将它们连接起来，可以在不影响生产科学的情况下进行实验。 FAB 提供与生产网络和专用测试台的可编程对等互连，允许实验者拓扑与生产网络相结合，从而极大地扩展了可利用基础设施的资源类型和用户的可能性。 FAB 为设施内的研究人员创建新的软件服务和工具，并与现有和不断发展的数据交付服务接口，以在全球范围内高效地移动和处理科学数据，并测试可扩展到海量的新颖数据分析方法。成功的衡量标准是由科学实验本身驱动的：更有效地处理来自欧洲核子研究中心实验的高能物理数据以及在南美洲和南极收集的宇宙微波背景数据；成功验证了共享智能城市传感器数据以进行城市规划以及建立全球专用 5G 网络的概念。与 FAB 相关的所有软件都将开源并发布在公开可用的存储库中：https://github.com/fabric-testbed/ 该奖项反映了 NSF 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。