CI-ADDO-EN: Scaling, Diversifying and Enriching Planet Lab

该项目正在支持和扩展PlanetLab研究测试平台。PlanetLab是一个1000节点的覆盖网络，覆盖全球500个站点和40多个国家。 它用于评估新的网络服务（例如，对等网络、内容分发、可缩放对象位置服务和自适应路由覆盖），以及观察因特网活动（例如，网络拓扑、路由动态、路径等待时间和带宽以及故障属性）。 在支持PlanetLab的过程中，该项目的重点是使研究社区能够为可用的计算和通信资源集增加多样性和规模。该项目还通过改进仪器来扩展PlanetLab。 我们的方法是利用目前用于运营PlanetLab的数据收集和分析设施，为PlanetLab用户提供更多价值。这有两个方面。首先，该项目将所有工具收集的原始数据提供给公众下载。 当然，数据可以有多种用途，而且似乎用于操作的相同数据可能对社区具有研究价值。 其次，该项目正在扩展仪器和相关工具集，以帮助用户回答他们在开发广泛分布的网络服务时所面临的各种问题：“我的服务有什么问题吗？如果是这样，原因是什么，我如何解决它？“我们已经投资建立了一套高效的数据收集设施和强大的分析工具，该项目的目标是看到它们解决尽可能多的问题。

This project is supporting and extending the PlanetLab research testbed. PlanetLab is a 1000-node overlay network spanning 500 sites and 40+ countries world-wide. It is used to evaluate new network services (e.g., peer-to-peer networks, content distribution, scalable object location services, and adaptive routing overlays), as well as to observe Internet activity (e.g., network topology, routing dynamics, path latency and bandwidth, and failures properties). In supporting PlanetLab, this project is focused on enabling the research community to add diversity and scale to the set of available computing and communication resources. The project is also extending PlanetLab with improved instrumentation. Our approach is to leverage the data collection and analysis facilities currently used to operate PlanetLab to provide more value to PlanetLab users. There are two aspects to this. First, the project is making the raw data collected by all tools available for public download. Of course data can have multiple uses, and it seems likely that the same data used for operations may be of research value to the community. Second, the project is extending the instrumentation and related toolset to help users answer the kinds of questions they face when developing a widely distributed network service: "Is there something wrong with my service? If so, what is the cause, and how do I fix it?'' Having already made the investment to build a set of efficient data collection facilities and powerful analysis tools, the goal of the project is to see them solve as many problems as possible.