Five to ten more years left before moving on to IPv6 and abandoning IPv4

In light of Huawei’s proposal to abandon the TCP / IP stack and switch to a completely new model for new IPs, which was not met with much enthusiasm, one cannot help but recall the existence the next generation IPv6, which in theory should serve as proof of the flexibility of the current generation of technology. Indeed, after a quarter century, a total transition to IPv6 has not yet taken place.

This transition should take place within the framework of the current model, because in November of last year, the free IPv4 addresses had run out. Now only returnable address pools are available in the “secondary” market, which often skirts the lines of legality. Ad hoc palliative measures such as CG-NAT or the simultaneous support of both versions of the protocol only complicate the work of the Network and hamper the process of moving away from IPv4.

There is no problem with the standard as such; it has been a mature technology for a long time. According to Google, IPv6 penetration is about a third, and in some countries approaching 50%. In a comment in The Register, RIPE NCC spokesman Marco Hogewoning noted that a further five to ten years could pass before IPv4 is fully abandoned. And there are a number of reasons for this. In particular, despite the fact that at least minimal support for the new protocol has long been built into the software and hardware, not everyone is ready to make the change as this will require investment in the updating, configuring, and optimizing of the entire associated infrastructure.

For example, the US government is trying to forcibly transfer government services to IPv6. The Pentagon received just such an instruction as early as 2003, but things have since stagnated there — the ministry is not ready to provide an estimate of the approximate cost of the transition, citing security issues. It can be assumed that the true reason is that IPv6 makes it possible to use “white” addresses on all devices and systems in general, and this leads to the need for additional protection measures. In general, it is clear that any inaccessibility of public services to end users and amongst the services themselves will lead to serious and embarrassing delivery problems, and also public services themselves are well-known for their creaking institutional inertia and in-grained resistance to change.

There is another aspect to consider, one not only and not so much of a technical nature, but of the stability of the Network segments. Providers and associated organizations are bound in a web of complex relationships pertaining to peering and traffic transit: who, how much, and to whom fees should be paid, if indeed they should be paid at all. Disputes arise both globally and locally. This directly affects the quality and stability of the connection. It may turn out that basic IPv4 and the alternative options may be more than adequate to maintain transit traffic flows, while IPv6 it is still very limited, and so the failure of even one of transit points is an unacceptable risk. To correct this situation, mass coordination is needed, a commitment to which not everyone is ready to make.

HOSTKEY provides additional IP-addresses individually (1–31) or in the following packages: 32, 64, 128, 256, 512 IPs in Russia and the Netherlands. Blocks can be routed via VLAN to the client’s server.

HOSTKEY also fully supports IPv6 on all of our platforms. If you would like to order IPv6 /64 block of IP addresses, please contact our technical support. IPv6 addresses are allocated free of charge.