Skip to Content
[CAIDA - Center for Applied Internet Data Analysis logo]
Center for Applied Internet Data Analysis > publications : papers : 2015 : ipv6_as_relationships
IPv6 AS Relationships, Clique, and Congruence
V. Giotsas, M. Luckie, B. Huffaker, and k. claffy, "IPv6 AS Relationships, Clique, and Congruence", in Passive and Active Network Measurement Workshop (PAM), Mar 2015, pp. 111--122.
|   View full paper:    Data Supplement    PDF    Related Presentation    |  Citation:    BibTeX   |

IPv6 AS Relationships, Clique, and Congruence

Vasileios Giotsas
Matthew Luckie
Bradley Huffaker
kc claffy

CAIDA, San Diego Supercomputer Center, University of California San Diego

There is increasing evidence that IPv6 deployment is maturing as a response to the exhaustion of unallocated IPv4 address blocks, leading to gradual convergence of the IPv4 and IPv6 topologies in terms of structure and routing paths. However, the lack of a fully-connected transit-free clique in IPv6, as well as a different economic evolution than IPv4, implies that existing IPv4 AS relationship algorithms will not accurately infer relationships between autonomous systems in IPv6, encumbering our ability to model and understand IPv6 AS topology evolution. We modify CAIDA‛s IPv4 relationship inference algorithm to accurately infer IPv6 relationships using publicly available BGP data. We validate 24.9% of our 41,589 c2p and p2p inferences for July 2014 to have a 99.3% and 94.5% PPV, respectively. Using these inferred relationships, we analyze the BGP-observed IPv4 and IPv6 AS topologies, and find that ASes are converging toward the same relationship types in IPv4 and IPv6, but disparities remain due to differences in the transit-free clique and the influence of Hurricane Electric in IPv6.

Keywords: ipv6, measurement methodology, routing, topology
  Last Modified: Wed May-3-2017 10:57:35 PDT
  Page URL: