CC* Networking Infrastructure: FAN4Science: Flow-Aware Networking for Data-Intensive Science

使对地方、地区和国家超级计算设施的访问民主化，对于研究人员实现他们日益数据密集型的科学目标至关重要。此外，研究与国家或全球研究团队非常合作，研究全球流行病、气候变化、智慧城市和探索宇宙的秘密等现象。这些目标可以通过研究和教育网络的互连网络高速连接到强大的计算平台来实现。FAN4Science项目汇集了网络工程、校园IT和区域网络提供商的专家，为中佛罗里达大学S的研究人员解决实现高吞吐量和灵活网络的研发挑战。该项目的科学价值是探索使用流感知网络方法来为研究人员增加校园网络的有效容量，同时也使计算机网络的创新研究成为可能。该项目(A)建设一个100Gbps的研究网络，以使目标研究实验室能够高速访问UCF的高级计算设施；(B)在关键位置安装性能监测和数据传输节点，以监测研究数据的移动；(C)通过为UCF研究和教学活动设置软件定义网络(SDN)来增强园区网络；以及(D)利用真实流量，实现实验性SDN研究，探索推测流量安装，以减少端到端延迟。该项目为UCF研究生提供了研究机会，并提供了安装真实网络和性能监控设备的实践学习体验。吸取的经验教训有可能演变为SDN控制的校园网络设计，可以处理常规的校园流量和研究流量。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Democratizing access to local, regional, and national supercomputing facilities is essential for researchers to meet their increasingly data-intensive science objectives. Furthermore, research is exceedingly collaborative with national or global research teams studying phenomena like global pandemic, climate change, smart cities, and exploring the universe's secrets. These objectives can be achieved through high-speed connectivity to powerful computing platforms through the interconnected web of research and education networks. The FAN4Science project brings together experts in network engineering, campus IT, and regional network providers to address research and development challenges of enabling a high-throughput and flexible network for University of Central Florida (UCF)'s researchers. The scientific merit of this project is to explore the use of flow-aware networking methods to increase effective capacity of a campus network for researchers while also enabling innovative research in computer networking. The project (a) builds a 100Gbps research network to enable targeted research labs to get high-speed access to UCF's advanced computing facility; (b) installs performance monitoring and data transfer nodes at key locations to monitor research data movement; (c) augments the campus network with a Software-Defined Networking (SDN) setup for UCF research and teaching activities; and (d) enables experimental SDN research, utilizing real traffic, that explores speculative flow installations to reduce end-to-end latency. The project offers research opportunities to UCF graduate students and hands-on learning experience with installing real networking and performance monitoring equipment. Lessons learned have the potential to evolve into an SDN-controlled campus network design that can handle regular campus traffic and research traffic.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.