NSF Award
0820944
In this SGER, BBN is proposing to engage in preliminary work on two untested and novel ideas. The first idea focuses on the challenge of how to generate and support realistic traffic flow on experimental, network infrastructure. Other network infrastructure typically relies on artificial traffic generation methods or user communities that are composed of the experimenters themselves. But, in order to provide experimenters with real traffic at scale, novel approaches and mechanisms need to be developed. If such mechanisms can be developed and real traffic is carried on the infrastructure, many hard problems remain. For example, how many and what types of opt-in mechanisms are actually needed? Will the infrastructure have enough ?user pull? (e.g., applications, capabilities, etc.) to attract users? What kind of guarantees can be given to the users ? in terms of privacy, reliability, security, etc? These questions can not currently be answered as new interdisciplinary knowledge needs to be generated. This SGER will allow for the development of a series of white papers on this topic, including 1) strategies to encourage opt-in, 2) applications that could drive user opt-in, 3) legal, ethical, privacy and security implications of user opt-in, 4) prospective user communities and methods for accessing them, and 5) the potential benefits of user opt-in for the commercial sector. The authors of these papers will meet at the Engineering Conferences to discuss and share their findings and how they might be implemented in an experimental network infrastructure.<br/><br/>The second novel idea revolves around broadening research engagement and interest in network infrastructure. It is essential that the widest range of scientists and engineers with research interests beyond networking and distributed systems be allowed to carry out their own experiments on novel, network infrastructure. Other disciplines that might be interested include the social sciences, economics, business, law, medicine, and alternative energy, for example. Such a broad purview helps to ensure that new network infrastructure will widely influence research and education. But it is essential that their requirements for experimental workflow, data access and measurement, for example, be generated now so that they can influence infrastructure design. If those requirements are not known early, some areas of science may be locked out from use of the infrastructure. It is critical that experts be engaged now to generate those requirements. This SGER will help to generate infrastructure requirements for network researchers in the social sciences, law, economics, and other disciplines interested in network science and engineering, which will allow for new network infrastructure to have the broadest possible impacts.
