It has long been recognized that the Internet lacks the flexibility to provide enhanced services. While proposals to augment the Internet abound, efforts to modify its infrastructure have essentially failed. As a response, overlay networking gained attention as a viable an approach to provide enhanced Internet services. Another response has been the push to define a new clean-slate Internet architecture. Network virtualization is emerging as a promising new central feature of the future Internet architecture. Interestingly, but perhaps not surprisingly, the use of overlay networking in the current Internet, as well as the potential deployment of network virtualization in a future Internet both result in what we dub as multi-layered networks. The goal of this project is to undertake a systematic investigation of this paradigm, focusing primarily on routing functions. The research aims to provide a unified understanding that will be applicable in both current and future Internet environments. The project also aims to articulate fundamental differences between current and future environments that will affect how we use or design multi-layered networks. A central concern is the interaction between routing protocols in the multiple layers.
Broader Impact: Since the project considers both the current and a future Internet, results from this project will more broadly help improve our understanding of network virtualization in a future Internet and how it can achieve its full potential. The PI is also committed to activities with additional broader impact including: development of modules based on our research for classes and continued emphasis on inclusion of under-represented groups within the research.