Design requirements of a global name service for a mobility-centric, trustworthy internetwork

The Internet's tremendous success as well as our maturing realization of its architectural shortcomings have attracted significant research attention towards clean-slate re-designs in recent times. A number of these shortcomings can be traced back to naming. The current Internet uses IP addresses to conflate identity and network location, which results in poor support for mobility and multihoming; vulnerability to hijacking and spoofing of addresses, etc. The Internet's name resolution infrastructure deeply embeds in its design the assumption of mostly stationary hosts and poorly satisfies the performance, security, and functionality demanded by modern mobile services. As a step towards addressing these shortcomings, we present the design of a global name service that forms a central component of the MobilityFirst, a clean-slate Internet architecture with mobility and trustworthiness as principal design goals. MobilityFirst relies on the global name service to cleanly separate identity from network location and to resolve identifiers to locations in a secure manner. More importantly, MobilityFirst capitalizes on the role of the name resolution infrastructure as a logically central, first point of contact to significantly enhance a number of network-layer functions such as supporting host and network mobility, multi-homed traffic engineering, content retrieval, multicast, and next-generation context-aware services. This paper identifies key challenges that must be addressed to realize such a vision and outlines the design of a distributed global name service that can resolve identifiers to dynamic attributes in a fast, consistent, and cost-effective manner at Internet scales.