ISP Driven Informed Path Selection

We are seeing the emergence of technologies severely challenging some assumptions that have driven the development of many Internet protocols and mechanisms. A first assumption is that (usually) one address is associated to each host. Also, the forwarding of packets is often exclusively based on the destination address. For this reason, there is usually a single path between one source (or client) and one destination (or server). Finally, the Internet was designed with the client-server model in mind assuming that many clients receive information from (a smaller number of) servers. During the last years, these assumptions have been severely challenged. For example, the LISP protocol described in the previous chapters relies on this path diversity. The client-server model does not correspond to the current operation of many applications. First, large servers are usually replicated and different types of content distribution networks are used to efficiently distribute content [FFM03, Lim, Aka]. Second, the proliferation of peer-to-peer applications implies that most clients also act as servers. This is currently creating several problems in many Internet Service Provider (ISP) networks [KRP05]. Client-server asymmetry does not hold anymore. Due to the transition from IPv4 to IPv6 many hosts will be dual-stack for the foreseeable future [CLH03]. Furthermore, measurements show that IPv4 and IPv6 do not always provide the same performances, even for a single source-destination pair [ZJUVM07]. This implies that to reach a destination supporting both IPv4 and IPv6, a source host can achieve better performance by selecting the stack that provides the best performance. However, today, this selection is based on simple heuristics. For instance, as highlighted by Matsumoto et al. [MFHK08], IPv6 is chosen prior to IPv4 in most of the dual-stack configurations although IPv4 still remains the best choice from a performance point of view in many environments [ZJUVM07]. We are thus heading towards an Internet that provides a set of potential paths between a source and a destination or a content. Obviously, any path in this set differs from the others as each path has its own performances,