Evolution of networking: Current problems and future directions

Abstract

Networking, as a piece of technology, seems to be on the brink of its second fundamental revolution. The first revolution was the move from wires to packets; a phenomenon that started in late 1950's, gained speed in the 1980's and 1990's, and seems to be coming to its conclusion right now. The second revolution, moving from packets to information, is only starting. Like the first revolution, the second revolution will bring forth a number of new types security problems, many of which we will even notice only later on. In this talk, we use the revolutions framework to discuss current and anticipated networking-related problems, both architectural and security related. Currently, the primary reasons behind the most pressing problems include loss of trust, surge of unwanted traffic, choking of the routing system, poor support for mobility and multi- homing, and lack of privacy and accountability. The attempts to alleviate these problems, including middle-boxes, moving connectivity to overlay networks, and virtualisation, are at best temporary. More fundamental chances are needed to bring forth long-lasting solutions. Hence, we conclude with a brief look at a few of the most interesting research directions going on at the time, illustrating some aspects of the envisioned information-centric networking paradigm.