A study on the distribution of active IPv6 addresses used by websites

Website IPv4 address distribution can be learned by scanning the entire IPv4 address space. However, the huge space of IPv6 addresses makes it difficult to collect and analyze active website IPv6 addresses. In this paper, we collect a dataset of nearly half a million IPv6 addresses visited by domain names through an experimental research network based on IVI, a stateless IPv4/IPv6 translation technique. As a result, a study was conducted on the dataset, which shows us the characteristics of geographic location, address type, ASN, and ports. Our result demonstrates that active website IPv6 addresses have obvious features especially in ASN. In addition, we compare different algorithms of mapping an IPv6 address into an IPv4 address.