NeTS: Medium: Collaborative Research: PacketLab: A Universal Measurement Endpoint Interface

packketlab是一个通用的互联网流量测量抽象，它降低了互联网测量基础设施的实验者和运营商面临的障碍。互联网测量对科学和社会产生了广泛的重大影响，从帮助我们识别僵尸网络等网络威胁，到为讨论互联网政策问题提供事实基础。这项工作旨在通过开发一个通用接口来简化网络测量并促进研究小组之间的资源共享，从而满足整个网络测量社区的需求。PacketLab将使共享测量资源变得更容易，这将使那些通常没有资源来开发和维护自己的测量平台的机构更容易获得这些资源。本项目旨在回答网络测量中的几个基本问题，包括：1)如何使用少量测量原语来表达测量实验；2)实验逻辑能否有效地移出进行测量的端点；3)测量平台运营者如何在鼓励资源共享的同时保证平台政策的一致性。PacketLab建立在两个关键思想之上。首先，它将测量逻辑从测量端点转移到单独的实验控制服务器，使每个端点成为轻量级数据包源或接收器。其次，它提供了一种方法来授权对测量端点的访问，同时保留对实验人员如何使用端点的细粒度控制，允许研究小组以很少的开销彼此共享测量基础设施。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。

PacketLab is a universal internet traffic measurement abstraction that lowers barriers faced by both experimenters and operators of internet measurement infrastructures. Internet measurement has had a major impact on science and society broadly, from helping us in identifying network threats like botnets, to providing a factual foundation for discussing internet policy issues. This work aims to serve the needs of the entire network measurement community by developing a universal interface that simplifies network measurement and facilitates resource sharing between research groups. PacketLab will make it easier to share measurement resources, and this will broaden access to these resources by institutions that would normally not have the resources to develop and maintain their own measurement platform. This project aims to answer several fundamental questions in network measurement including 1) how to express measurement experiments using a small number of measurement primitives; 2) whether experiment logic can be effectively moved off the endpoint conducting the measurement; and 3) how measurement platform operators can ensure conformance with their platform policies while encourage resource sharing. PacketLab is built on two key ideas. First, it moves measurement logic away from the measurement endpoint to a separate experiment control server, making each endpoint a lightweight packet source or sink. Second, it provides a way to delegate access to measurement endpoints while retaining fine-grained control over how endpoints are used by experimenters, allowing research groups to share measurement infrastructure with each other with little overhead.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.