Conceptual model for comparison of IPv6 ISPs based on IPv4 traffic profiles

Abstract

The slow adoption of IPv6 protocol by organizations has led to internet service providers (ISP) investing minimal resources in setting up IPv6 Internet connectivity, ultimately resulting in sub optimally tuned IPv6 Internet routing fabric. The altogether different IPv6 packet header with more bits for the IP address, no header checksum, and no intermediate fragmentation features, are expected to give a palpably different experience to the users. With different levels of expertise and different connectivity links to the Internet, available with different ISPs, responses of the destined networks may also vary for each ISP. Therefore, it is worth to investigate ISP wise IPv6 Internet responses from an organization user’s point of view. In this paper, we present a research methodology to estimate IPv6 responses to the users of an organization for destinations present in current IPv4 traffic profile, without actually accessing the IPv6 network. We use publicly available centre for applied internet data analysis (CAIDA) IPv6 topology and autonomous systems (AS) links datasets to analyse and infer the intermediate AS level responses, for estimating responses between source and destination networks. As a case study, we take three Indian ISPs and apply the methodology to compare ISP wise responses for a network of about 2000 nodes. The results of the same are presented in this paper. We conclude that considering all other factors to be the same, the best possible quality of experience (QoE) for the Internet users of the organization, can be achieved by smart selection of the IPv6 ISP.