CNS Core: Medium: Slack-Aware Networking

今天的互联网正越来越多地向可能不立即需要数据的通信方向发展：机顶盒或智能手机应用程序预先获取内容，物联网（IoT）设备将其数据上传到云端，或者数据中心进行地理分布式复制。这项研究假设，这种通信可能会有内置的“松弛”，也就是说，另一个端点可能不会立即需要数据包，并且可以容忍几秒钟甚至几分钟和几小时的额外延迟。Slack允许网络实现“松弛感知网络”——网络可以利用松弛信息来降低网络的成本和能源需求，并为最终用户提高性能、可用性和成本。本项目将探讨如何使slack成为未来互联网的一流概念所面临的挑战和机遇。它将探讨SAIN的设计，这是一个在应用层面和网络内部明确支持松弛的松弛感知互联网。SAIN将允许应用程序向网络表达其松弛需求，而网络将在系统的各个层面（传输、路由、转发）使用松弛感知优化，为网络和服务提供商以及最终用户提供好处。SAIN的潜在效益将在广泛的应用中进行评估，包括视频点播、物联网和地理分布式复制。总的来说，这项研究将在网络和系统研究以及网络和人机交互（HCI）的交叉领域做出关键贡献。具体来说，通过用户研究和应用程序分析，拟议的研究将提供关于流行应用程序的使用情况以及它们可以容忍的闲置量的见解。它还将在网络堆栈的多个层引入新功能，包括用于解耦发送方和接收方的存储抽象、松弛感知路由和转发机制、松弛感知端到端数据服务，以及应用程序和网络之间的新松弛感知接口。拟议的研究将与塔夫茨大学PI提供的课程紧密结合，并将使本科生、妇女和URM学生参与综合研究和教育。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。

Today's Internet is increasingly moving towards communication where data may not be needed immediately: set-top-boxes or smart-phone applications pre-fetching content, Internet of Things (IoT) devices uploading their data to the cloud, or data centers doing geo-distributed replication. This research posits that such communication may afford to have ``slack'' built into it, i.e., packets may not be immediately required by the other endpoint and can tolerate additional delay of few seconds to even minutes and hours. Slack allows the network to do ``slack-aware networking” -- the network can leverage slack information to reduce the cost and energy requirements of the network, and to improve performance, availability and cost for end-users. This project will investigate the challenges and opportunities present in making slack a first class concept of the future Internet. It will explore the design of SAIN, a slack-aware Internet with explicit support for slack at the application level as well as inside the network. SAIN will allow applications to express their slack requirements to the network while the network will use slack-aware optimizations at various levels of the system (transport, routing, forwarding), providing benefits to the network and service providers, as well as end-users. The potential benefits of SAIN will be evaluated on a wide range of applications, including Video-on-Demand, IoT, and Geo-Distributed Replication. Overall, this research will make key contributions in networking and systems research as well as at the intersection of networking and human-computer-interaction (HCI). Specifically, through user studies and application analysis, the proposed research will provide insights about usage of popular applications and the amount of slack they can tolerate. It will also introduce new functionality at multiple layers of the network stack, including storage abstractions to decouple the sender and receiver, slack-aware routing and forwarding mechanisms, slack-aware end-to-end data services, and a new slack-aware interface between the application and the network. The proposed research will be tightly integrated with the courses offered by the PI at Tufts University, and will engage undergraduate, women and URM students in integrative research and education.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.