Enabling ICN in IP networks via DNS Extensions: A Measurement Study
ICN promises benefits to users and service providers along several dimensions, including lower response time, nearest-replica routing, intrinsic content integrity without external network-level indicators, location-independent naming and improved support for mobility. However, these benefits come at non-trivial cost as many ICN proposals call for adding significant complexity to the network. Moreover, many open questions regarding content naming, caching, routing, security, discovery and scalability remain to be solved before any proposal can be deployed at Internet scale. It has been proposed that some of the benefits of ICN can already be achieved through a number of modifications of the Domain Name System (DNS), enabling its servers to resolve Named Data Object (NDO) names in addition to domain names. Given that the global DNS system, especially at the higher root and top-levels, already experiences significant load levels we seek answers to the following questions: 1) How many additional, NDO-denoting records need to be stored at the authoritative levels? 2) Which load is expected at caching DNS resolvers, and which cache size is required? 3) How many additional resources are required where in the DNS hierarchy? To answer some of these questions we performed case studies of DNS traffic based on URL traces. Based on the results we give and estimation on the resources required for the modified DNS to perform at scale.