This project (in collaboration with Robert Beverly, Naval Postgraduate School) seeks to characterize the status of IPv6 deployment while simultaneously advancing the state-of-the-art in network measurement science and technology.
Funding source: NSF CNS-1111449. Period of performance: May 1, 2012 - April 30, 2016.
Broad industry awareness of IPv4 address scarcity has driven universal support for IPv6 in most modern operating systems and network equipment. Although the uptake of IPv6 is still slow, its adoption renders many existing methods to measure and monitor the Internet ineffective. In this project, we improved the fidelity, scope, and usability of IPv6 measurement technology and collected vast amounts of data on IPv6 topology and performance. We expanded our Ark measurement infrastructure, deploying more monitors capable of IPv6 probing all over the world. We designed and tested measurement primitives for adaptive and intelligent probing, boosting the efficiency of IPv6-scale topology measurements. We developed, implemented, and validated the first Internet-scale alias resolution technique and the first Internet-scale AS routing relationship inference technique for IPv6. We also created a completely new system for bulk lookups of DNS mappings for IPv4 and IPv6 addresses.
We also developed and applied novel analysis methodologies. We investigated the AS-level congruency of IPv4 and IPv6 paths in the Internet and found that more than 60% of the current IPv4 and IPv6 AS-paths were non-congruent at the AS-level. Multi-Path TCP can exploit this condition to improve application performance in addition to providing resilience. We compared stability, and RTT performance of IPv4 and IPv6 paths and found that in both control and data plane experiments, IPv6 exhibited less stability than IPv4. We also found evidence of correlated performance degradation over IPv4 and IPv6 caused by shared infrastructure.
Finally, we correlated address allocation patterns with observable traffic and geopolitical factors examining per-country allocation and deployment rates through the lens of the annual "Day in the Life of the Internet" snapshots collected at the DNS roots by the DNS Operations, Analysis, and Research Center from 2009-2014. Our interactive data analysis is in the page Dataset Comparison: IPv4 vs IPv6 traffic seen at the DNS Root Servers.
Our analysis methodologies and research results provide much needed insights to network operators, governments, and Internet policy makers about the status, progress, challenges, and future of the IPv6 in the global Internet.
Intellectual Merit. The project advanced the state-of-the-art in network measurement and technology while providing empirically grounded understanding of the crucial architectural transition that the Internet is currently undergoing. Our rigorous approach to the study of IPv6 deployment not only represents a compelling case study of transformations in the current Internet, but is applicable to technology transfer challenges in other domains as well.
Broader Impact. We widely disseminated our results to research, commercial, and government sectors, helping to inform communications and technology policies. We published 12 papers, organized and hosted 4 workshops, and released 3 ongoing and 2 one-time snapshot data sets. In order to advance community interest and involvement in IPv6 research, all our IPv6 related data sets are publicly available for download.