The earliest predecessor to the commercial Internet of today was ARPANET, a packet switched computer network developed by the US Defense Department's Advanced Research Projects Agency. Designed to withstand a nuclear attack, ARPANET utilized a deurbanized, decentralized, and distributed network topology. As ARPANET gradually evolved into NSFNET and eventually the commercial Internet, increasing traffic, demands for interconnection, and growing private interests required the movement from a distributed network topology to a more economically viable network configuration, hub-and-spoke. Although transmission speeds and capacities of today's commercial Internet clearly surpass those of its predecessors, the economics of network survivability and reliability have also become more relevant. With thousands of businesses, corporations, universities, and governments relying on the Internet for day-to-day functions, major disruptions in service have the potential to be economically catastrophic. This paper explores the network topology of the commercial Internet, with a focus on network survivability. GIS based approaches are used to simulate both nodal and link failure in the US commercial backbone system in order to assess potential impacts. Results suggest that many of the larger metropolitan benefit from robust network infrastructure, while smaller cities are more prone to service disruptions.
ASJC Scopus subject areas
- Computer Networks and Communications
- Electrical and Electronic Engineering