NeTS: Small: Creating an Evolvable, Diverse, and Dynamic Internet

NeTS：小：创建一个可进化、多样化和动态的互联网

• 批准号: 1817115
• 负责人: Scott Shenker
• 金额: $ 50万
• 依托单位: International Computer Science Institute
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-10-01 至 2022-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1817115&HistoricalAwards=false
• 关键词: NeTS; Small; Creating; Evolvable; Diverse

The Internet ushered in a new era of communication, and it has engendered an ever-growing set of applications that have transformed our lives. It is remarkable that all this has taken place with an Internet architecture that has remained unchanged for over forty years. While much research has been done on promising new Internet architectures, there is little hope for their adoption because the pain of transitioning to a new architecture is too great. When it comes to Internet architecture, we know much more about where we want to go, than we do about how to get there. This project looks not at what new architectures we should create, but how we could more easily deploy them. The goal is to create an Internet that is more evolvable and dynamic, so that what has served us so well over the past forty years can evolve to meet our needs over the next forty and more. Some view the forty years of architectural stagnation as inevitable. After all, it has long been a central tenet that the Internet needs a "narrow waist" at the internetworking layer (L3), a single uniform protocol adopted by everyone; given this assumption, changing this layer is inevitably hard. This project rejects this dogma, attributing the requirement of a narrow waist to a fundamental flaw in the original Internet architecture: the conflation between the interdomain dataplane and the intradomain dataplanes. As an alternative, this project separates these two dataplanes. Furthermore, in copying the Internet?s use of layering, in which L3 is an overlay on L2, and L2 is an overlay on L1, this project implements the interdomain dataplane (referred to as L3.5) as an overlay on top of the intradomain dataplane (L3). This decouples the two dataplanes, allowing them to evolve independently. This project uses this approach to create an architectural framework that enables all clean-slate architectures to be incrementally deployable and coexist, making the Internet fully extensible. Moreover, rather than being some radical clean-slate architecture that could never be deployed, the framework itself is incrementally deployable.Enabling architectural extensibility could greatly change the nature of the Internet. Today, an Internet architecture must try to meet all the needs of all applications. Once the Internet becomes fully extensible, with an ever-changing set of coexisting architectures, each architecture can be designed to more fully meet only some of the needs of some applications. This would hopefully lead to a broader ecosystem of architectures which, in their union, can better meet all the needs of all applications.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.

互联网开创了一个新的通信时代，它产生了一套不断增长的应用程序，改变了我们的生活。值得注意的是，所有这一切都发生在一个四十多年来一直保持不变的互联网架构上。虽然已经对有前途的新Internet架构进行了大量研究，但采用它们的希望不大，因为过渡到新架构的痛苦太大了。当涉及到互联网架构时，我们对我们想去哪里的了解要比我们对如何到达那里的了解要多得多。这个项目关注的不是我们应该创建什么样的新架构，而是我们如何更容易地部署它们。我们的目标是创造一个更可进化和更有活力的互联网，这样在过去40年里为我们服务的东西就可以进化，以满足我们在未来40年甚至更长时间的需求。有些人认为四十年的建筑停滞是不可避免的。毕竟，长期以来，互联网在互联网层（L3）需要一个“窄腰”一直是一个中心原则，每个人都采用一个统一的协议;考虑到这个假设，改变这一层不可避免地很难。这个项目拒绝了这个教条，将窄腰的需求归因于原始Internet架构中的一个根本缺陷：域间数据平面和域内数据平面之间的混淆。作为替代方案，该项目将这两个数据平面分开。此外，在复制互联网？L3是L2上的覆盖层，L2是L1上的覆盖层，本项目将域间数据平面（称为L3.5）实现为域内数据平面（L3）之上的覆盖层。这将两个数据平面合并，允许它们独立发展。这个项目使用这种方法来创建一个架构框架，使所有干净的架构能够增量部署和共存，使互联网完全可扩展。此外，框架本身是可增量部署的，而不是永远无法部署的激进的白板架构。启用架构可扩展性可以极大地改变Internet的本质。今天，互联网架构必须设法满足所有应用程序的所有需求。一旦互联网变得完全可扩展，并且有一组不断变化的共存架构，每个架构都可以被设计为更完全地满足某些应用程序的某些需求。这个奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Making edge-computing resilient

使边缘计算具有弹性

• DOI: 10.1145/3419111.3421278
• 发表时间: 2020
• 期刊: SOCC 2020
• 作者: Harchol, Yotam;Mushtaq, Aisha;Fang, Vivian;McCauley, James;Panda, Aurojit;Shenker, Scott
• 通讯作者: Shenker, Scott

From cloud computing to sky computing

• DOI: 10.1145/3458336.3465301
• 发表时间: 2021-06
• 期刊: Proceedings of the Workshop on Hot Topics in Operating Systems
• 作者: I. Stoica;S. Shenker

Revitalizing the Public Internet By Making it Extensible

• DOI: 10.1145/3464994.3464998
• 发表时间: 2021-04-01
• 期刊: ACM SIGCOMM COMPUTER COMMUNICATION REVIEW
• 影响因子: 2.8
• 作者: Balakrishnan, Hari;Banerjee, Sujata;Zegura, Ellen
• 通讯作者: Zegura, Ellen

Bertha: Tunneling through the Network API

• DOI: 10.1145/3422604.3425927
• 发表时间: 2020-11
• 期刊: Proceedings of the 19th ACM Workshop on Hot Topics in Networks
• 作者: Akshay Narayan;Aurojit Panda;Mohammad Alizadeh;H. Balakrishnan;A. Krishnamurthy;S. Shenker

On the Future of Congestion Control for the Public Internet

• DOI: 10.1145/3422604.3425939
• 发表时间: 2020-11
• 期刊: Proceedings of the 19th ACM Workshop on Hot Topics in Networks
• 作者: Lloyd Brown;Ganesh Ananthanarayanan;Ethan Katz-Bassett;A. Krishnamurthy;Sylvia Ratnasamy;Michael Schapira;S. Shenker