NSF Workshop on Programmable Networks

可编程网络——其功能可以由网络运营商改变的网络——正在兴起，从根本上改变了计算机网络的设计、建造和管理方式。可编程网络一直是网络研究的圣杯，这种设备（如网络接口卡和交换机）以及新的高级网络应用程序正在进入市场。由于需要提供快速的数据包处理，可编程网络设备的高速内存有限，对每个数据包只能进行少量的计算。该奖项支持一个解决该领域重大挑战的研讨会：重新思考网络的完整软件堆栈，从开发新的算法和数据结构，以有效地实现有限资源的网络应用，到开发高级编程框架，以简化编程。这需要多个研究团体的努力：网络、算法、编程语言和计算机体系结构。本次研讨会旨在将这些不同的研究团体聚集在一起，调查可编程网络的现状，确定未来的关键研究挑战，并交换意见。主要目标是激发从事可编程网络研究的不同社区之间的讨论，并扩大研究可编程网络的研究社区。作为研讨会的具体成果，参与者将制作一份报告，该报告将作为ACM SIGCOMM在线杂志《计算机通信评论》的社论发表。本报告将集中总结研讨会的讨论，概述该领域未来研究的一系列研究主题，并就基础设施支持提出建议，以帮助可编程网络的研究。该奖项反映了美国国家科学基金会的法定使命，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。

Programmable networks---networks whose functionality can be changed by a network operator---are emerging to fundamentally change the way computer networks are designed, built, and managed. Programmable networks have been a holy grail for networking research, and such devices (such as network interface cards and switches) as well as new advanced network applications are coming to market. Due to the requirement of providing fast packet processing, programmable network devices have a limited amount of high-speed memory and can perform only a small amount of computation on every packet. This award supports a workshop addressing a grand challenge in this space: to rethink the full software stack of networks, from developing new algorithms and data structures to efficiently realize network applications with limited resources, to developing high-level programming frameworks to simplify programming. This requires the efforts of multiple research communities: networking, algorithms, programming languages, and computer architecture.This workshop aims to bring together these different research communities to survey the state of the art in programmable networks, to identify key research challenges going forward, and to exchange ideas. The primary goal is to stimulate discussion between different communities engaged in research on programmable networks and to broaden the research community working on programmable networks. As a concrete output from the workshop, participants will produce a report that will be published as an editorial in ACM SIGCOMM's online magazine, Computer Communication Review. This report will cohesively summarize the workshop's discussions, outline a set of research themes for future research in the area, and make recommendations on infrastructure support to aid research in programmable networks.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.