CC* Integration-Large: Democratizing Networking Research in the Era of AI/ML

CC* 大型集成：AI/ML 时代的网络研究民主化

• 批准号: 2126327
• 负责人: Arpit Gupta
• 金额: $ 99.99万
• 依托单位: University of California-Santa Barbara
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2126327&HistoricalAwards=false
• 关键词: CC; Integration; Large; Democratizing; Networking

The emerging area of self-driving networks provides network administrators at campus networks to automate most network-management tasks. Such an automation ensures that the network remains performant and reliable amidst various disruptions, while requiring minimal interventions from the network administrators. However, making significant contributions to self-driving network research requires developing artificial intelligence (AI) and machine learning (ML)-based tools and demonstrating that they work in practice. Unfortunately, in stark contrast to their counterparts in industry, most academic researchers have neither access to the proper data for developing learning-based tools nor have properly instrumented testbeds for road-testing the resulting tools in realistic settings. This collaborative project brings together investigators from the University of California-Santa Barbara, University of Chicago, and NIKSUN Inc., to investigate how to use campus networks to overcome barriers to self-driving network research. First, it will deploy packet-processing pipelines at two campus networks to collect the proper network data at scale without compromising user privacy. It will then strategically place programmable network devices at campus networks to safely road-test newly developed learning models in production settings. Finally, it will illustrate the capabilities enabled by these newly-instrumented campus networks for developing, evaluating, and road-testing new learning models with different use cases.This project is intended to seed a community effort that uses campus networks as vehicles for democratizing self-driving networks research, improving its transparency through reproducibility, and ensuring its success in practice by establishing trust. As such, it promises to be transformative not only for the network community as a whole but also for different campus network stakeholders (e.g., campus IT). Fully leveraging these campus networks' dual role as data source and testbed and seamlessly integrating it into the university's engineering curriculum suggests radically new approaches to teaching, training, and educating engineering students in the era of AI and ML.The project information will be maintained at: https://democratize-netai.cs.ucsb.edu/.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.

自动驾驶网络的新兴领域为校园网络提供网络管理员，以使大多数网络管理任务自动化。这样的自动化可确保在各种中断中保持网络的性能和可靠性，同时需要网络管理员的最小干预措施。但是，为自动驾驶网络研究做出重大贡献需要开发人工智能（AI）和基于机器学习（ML）的工具，并证明它们在实践中起作用。不幸的是，与行业中的同行形成鲜明对比的是，大多数学术研究人员既无法访问适当的数据来开发基于学习的工具，也没有适当的仪器测试台，用于在现实环境中进行对所得工具进行道路测试。这个合作项目汇集了加利福尼亚大学圣塔芭芭拉大学，芝加哥大学和尼克森公司的调查人员，以调查如何使用校园网络克服自动驾驶网络研究的障碍。首先，它将在两个校园网络中部署数据包处理管道，以大规模收集适当的网络数据，而不会损害用户隐私。然后，它将在校园网络上策略性地将可编程的网络设备放置在生产环境中的新开发的学习模型。最后，它将说明这些新启发的校园网络具有与不同用例的开发，评估和进行道路测试的新学习模型所启用的能力。该项目旨在播种社区努力，将校园网络用作民主化自动驾驶网络研究的工具，通过重新启动性来提高其透明度，并在实践中建立成功。因此，它有望不仅对整个网络社区，而且对不同的校园网络利益相关者（例如校园IT）具有变革性。充分利用这些校园网络作为数据源的双重作用，并将其无缝整合到大学的工程课程中，这表明在AI和ML时代，对工程学的学生进行了根本性的新方法。项目信息将维持在以下方面：认为值得通过基金会的智力优点和更广泛影响的评论标准来评估值得支持。

项目成果

A NetAI Manifesto (Part I): Less Explorimentation, More Science

NetAI 宣言（第一部分）：更少的探索，更多的科学

• DOI: 10.1145/3626570.3626609
• 发表时间: 2023
• 期刊: ACM SIGMETRICS Performance Evaluation Review
• 作者: Willinger, Walter;Gupta, Arpit;Jacobs, Arthur S.;Beltiukov, Roman;Ferreira, Ronaldo A.;Granville, Lisandro
• 通讯作者: Granville, Lisandro

Estimating WebRTC Video QoE Metrics Without Using Application Headers

• DOI: 10.1145/3618257.3624828
• 发表时间: 2023-06
• 期刊: Proceedings of the 2023 ACM on Internet Measurement Conference
• 作者: Taveesh Sharma;Tarun Mangla;Arpit Gupta;Junchen Jiang;N. Feamster

SPG: Structure-Private Graph Database via SqueezePIR

• DOI: 10.14778/3587136.3587138
• 发表时间: 2023-03
• 期刊: Proc. VLDB Endow.
• 作者: Ling Liang;Jilan Lin;Zheng Qu;Ishtiyaque Ahmad;Fengbin Tu;Trinabh Gupta;Yufei Ding;Yuan Xie

PINOT: Programmable Infrastructure for Networking

PINOT：可编程网络基础设施

• DOI: 10.1145/3606464.3606485
• 发表时间: 2023
• 期刊: ACM
• 作者: Beltiukov, Roman;Chandrasekaran, Sanjay;Gupta, Arpit;Willinger, Walter
• 通讯作者: Willinger, Walter

A NetAI Manifesto (Part II): Less Hubris, more Humility

NetAI 宣言（第二部分）：少一点傲慢，多一点谦虚

• DOI: 10.1145/3626570.3626610
• 发表时间: 2023
• 期刊: ACM SIGMETRICS Performance Evaluation Review
• 作者: Willinger, Walter;Gupta, Arpit;Beltiukov, Roman;Guo, Wenbo
• 通讯作者: Guo, Wenbo