CC* Integration: Rutgers University Next-Generation Edge Testbed (RU-NET)

随着大数据、深度学习、数据挖掘、云服务和网络安全等领域的最新进展，以及许多新的消费和科学设备，计算的技术格局正在以惊人的速度发生变化。研究型大学，尤其是IT支持，面临着在这种环境中保持敏捷的挑战，因为它们必须跟上研究和用户社区的需求。为了支持从事这些技术的研究人员，该项目寻求支持两个关键的以网络为中心的需求：在网络边缘整合用户拥有的设备，以及在网络上的数据收集节点和数据处理节点之间传输大量数据。该项目将设计、实施、测试和利用一个新的研究试验床--罗格斯大学下一代边缘测试床(RU-NET)。RU-Net是一个校园范围的平台，旨在简化新的和新兴的边缘技术的部署，同时保持校园网络的稳定性。RU-Net将允许将网络边缘实验室中的用户拥有的设备与罗格斯大学的三个园区网络集成到一个统一的100Gbit/S网络中，该网络使用可编程交换机和主机设备进行管理。RU-Net的设计是多层次的，由研究人员的应用程序、RU-Net拥有的物理资源和编排软件组成，将应用程序映射到资源并隔离不同的研究应用程序。RU-Net将充当催化剂，将罗格斯大学的研究人员定位在科学发现的前沿，并改进计算机和网络基础设施，使整个大学社区受益。该项目计划开发课程材料，教育本科生和研究生了解以网络为中心的新型技术，培训罗格斯大学不同学生群体中的学生。该项目还计划与正在规划类似分布式计算环境的学术机构和企业社区分享最佳实践，并有可能将该项目作为未来类似工作的模板。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。

The technological landscape of computing is changing at a phenomenal rate with recent advances in big data, deep learning, data mining, cloud services, and cybersecurity, along with many novel consumer and scientific devices. Research universities, especially IT support, are challenged to be agile in this landscape, as they must keep pace with the needs of the research and user community. To bolster researchers working on these technologies, this project seeks to support two key network-centric requirements: incorporating user-owned devices at the edge of the network and moving bulk data between data collection nodes and data processing nodes over the network.This project will design, implement, test, and utilize a novel research testbed, the Rutgers University Next-Generation Edge Testbed (RU-NET). RU-NET is a campus-wide platform designed to simplify the deployment of new and emerging edge technologies, while preserving the stability of the campus network. RU-NET will allow integrating user-owned devices in laboratories at the edge of the network with Rutgers's three campus networks into a unified 100 Gbit/s network managed using programmable switches and host devices. The RU-NET design is multi-layered, constituting researchers' applications, RU-NET-owned physical resources, and orchestrator software mapping applications to resources and isolating different research applications.RU-NET will serve as a catalyst to position Rutgers University researchers at the forefront of scientific discovery and to improve compute and network infrastructure that benefits the University community as a whole. The project plans to develop course material to educate undergraduate and graduate students about novel network-centric technologies, training students from Rutgers's diverse student body. The project also plans to share best practices with academic institutions and enterprise communities that are planning similar distributed computing environments, with the possibility of this project serving as a template for similar future efforts.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.