An agent-based model of IPv6 adoption

Despite having been proposed more than 20 years ago, IPv6 deployment has been very slow. The imminent depletion of IPv4 address space has recently motivated stakeholders to actively promote IPv6. These efforts, however, have only led to a relatively modest increase in the overall uptake. This outcome is expected given the complexity of the adoption landscape and the involved economic intricacies. Aiming to offer better insights into this process, we present the first data driven computational cost centric model of IPv6 adoption. Our model is grounded in empirical data yet parsimonious that is it focuses only on factors that are key to the modelled transition. We validate our model using historical snapshots of addresses allocations and then use it to explore a set of what if scenarios. Our findings paint a bleak picture for IPv6 adoption, predicting it to be decades away.