IPv6 network infrastructure and stability inference

Abstract

IPv6 deployment is increasing as IPv4 address allocations near exhaustion. Many large organizations, including the Department of Defense (DOD), have mandated the transition to IPv6. With the transition to IPv6, new techniques need to be developed to accurately measure, characterize, and map IPv6 networks. This thesis presents a method of profiling the uninterrupted system availability, or uptime, of IPv6 addressable devices. The techniques demonstrated in this study infer system restarts and the operational uptime for IPv6 network devices with a specific focus on IPv6 routers on the Internet. Approximately 50,000 IPv6 addresses were probed continuously from March to June 2014, using the Too Big Trick (TBT) to induce the remote targets to return fragmented responses. By evaluating the responses, the uptime for approximately 35% of the IPv6 addresses can be inferred.