The IPv4 to IPv6 transition: a novel self-balanced lossless translation model with dynamic address mapping

Since the next generation Internet protocol, IPv6, has been up and running for some years, the transition from the previous one, IPv4, has been strongly recommended, commonly due to a staggering increase in addresses. However, both coexist, although lacking the desired interoperability, which is useful for many purposes as it allows new applications to connect to services that are not ready for IPv6. In light of this initial statement, it emerges that what is currently being experienced is properly a ”transition phase”, even longer than what would have been expected, meaning that at the time is reasonable to believe that the two IP protocols will continue to be on the network for years, at least until IPv4 is no longer needed, certainly thanks to new implementations increasingly favorable to usability for businesses and users. First, technology needs time to develop and implement, both from the hardware and application point of view, setting a solid foundation for the new protocol. Secondly, thinking about the cost that companies would have to face in terms of human and economic resources, it is easy to see why this latest generation protocol has not established itself as a new standard. Thirdly, and as a consequence of what has just been said above, users in particular do not seem to be very well informed about it, with some exceptions, perhaps not so unexpected, for some emerging countries. Regardless, there are currently three mechanisms, called dual-stack, translation and tunneling, which carry out this task artificially, that is to say, the one relating to the planned and possible interoperability, on the understanding that it is clear that there is considerable room for improvement in terms of implementation and usability. For example, dual-stack provides a scalable and available network environment, while the translation mechanism allows native IPv6 and IPv4 nodes and applications to communicate specifically with each other. In spite of that, on the one hand, translators in particular are a good solution for companies, especially the smallest ones, by allowing the necessary adaptation period, i.e. by avoiding, as it has been pointed out, onerous costs in terms of implementation. On the other hand, on the point of view of the end users, the existing methodologies involve for the most part or a certain degree of knowledge, or additional charges, or of the initial study. In essence, they are not completely user friendly. Instead, in a wide-ranging perspective, what could encourage the spread of IPv6 is that many mobile networks, including 3G, have been designed on the basis of the IPv6 protocol standard, which means that their increasing use could also prove to be another valuable prerequisite for promoting the use of IPv6, as they are designed to offer high standards of end-to-end security and have significant support for wireless and mobile devices in general. In this regard, the massive adoption of technological devices in emerging countries, such as China, India and Arabia, to name but a few, where many are already working with IPv6, so much so that they seem increasingly willing to promote its popularity in the near future, could mean an important breakthrough. Besides, for several years, the most popular operating systems such as Windows, Linux and Solaris, just to name a few, have already been fully integrated with IPv6, as well as browsers, FTP clients and servers and almost all multimedia applications. Although many operating systems already use IPv6, there are still too many applicative requirements designed for IPv4 that support sockets (interface provided by the operating system for network communication), just considering that in Linux almost 90% of applications have this problem. Finally, another impediment, albeit unintentional, that has contributed to extend by far the period of time for the desirable global consolidation of IPv6, is represented by the translators themselves and the other related methodologies, previously only briefly listed but which in this thesis will be exhaustively illustrated, and technically described in detail, in their advantages and disadvantages. In the view of all these contingencies, even conflicting ones, it is clear that there is a real perspective, as well as a need closely linked to the expansion of the Internet network, in which the transition phase will have to be concluded. The key issue is therefore the one initially mentioned, that concerning interoperability. In fact, the main issue is that the IPv4 and IPv6 nodes cannot interact natively, and that currently to overcome this problem there are palliatives, albeit of good efficiency, but sometimes complicated to understand, other times expensive, and in any case there are limits that today represent the real challenge in terms of implementation and usability. The aim of this thesis is to propose a novel system of lossless and self-balanced translators designed in order to achieve the desired interoperability and scalability along with dynamic address mapping, thus solving the problem of high-level protocol failure without Application Layer Gateway (ALG). Finally, a proof of concept, through a prototype called DirecTo, is shown.