High-frequency mapping of the IPv6 Internet using Yarrp

Abstract

Both the number of hosts using Internet Protocol version 6 (IPv6), and the volume of IPv6 traffic, has increased exponentially since 2012.With this adoption, the IPv6 routed infrastructure becomes an increasingly important component of global critical infrastructure and network policy. Unfortunately, the tools and techniques used to perform active network topology discovery were designed for Internet Protocol version 4 (IPv4), leading to a potentially opaque view of the IPv6 Internet. In this thesis, we extend nascent work on stateless high-speed IPv4 active topology probing to develop a new IPv6 traceroute method Yelling At Random Routers Progressively version 6 (Yarrp6). Yarrp6 randomly permutes the set of IPv6 targets and hop counts to distribute load, thereby helping to avoid IPv6 response rate limiting. Further, we encode state in the IPv6 payload to permit Yarrp6 to both match responses with probes and use different probe transport protocols. Via active experimentation on the public IPv6 Internet, we compare the results obtained from Yarrp6 against the current state-of-the-art IPv6 topology mapping tool. We show that Yarrp6 can discover topology at more than an order of magnitude faster than previously possible. Finally, we conduct a study of the effect of transport layer protocol on forward Internet Protocol (IP) path inference to determine what protocol is best used for active IPv6 topology discovery.