Gemeric IPv6 Address Planing and Allocation Rules

IPv6 Address Planing for An ISP Network

Introduction

In the last article, we disccussed on the IPv6 implementation around the world. The addressing space of widely deployed Internet Protocol version (IPv4) had been depleted with continuous massive growth of IP based devices and applications over internet. Presently most of countries are planning and in process to adopt new version of Internet Protocol version 6 (IPv6) in their networks and applications for communication of a huge range of smart devices in near future. There are big challenges in migration from IPv4 to IPv6. Link is given in reference part.

This article covers a few rules to consider during design an IPv6 address plan. There are several RFCs that have been composed that examine IPv6 addresses. Few have been mentioned, for examples RFC 4291 and 4193, which characterize the IPv6 address architecture. So before diving into the point of internet identified with an IPv6 addressing plan. It is important to understand the basic rules when building an IPv6 address plan.

IPv6 address consist of 128 bits and these bits are divided into two portions such as “network number” and “host number”. Where network number of 64-bits is more specifically separated into the global unicast routing prefix and the subnet ID. The initial three bits of any all-around routable address are set to 001. Whereas it is characterized further 45 bits in global routing prefix. Remaining 16 bits are subnet ID. And the rest 64 bits of the prefixes are held for the interface ID.

Figure 1 IPv6 global routing prefix

Figure 2 IPv6 global routing prefix

As shown in figure 1 and figure 2 for usable bits of the global unicast routing prefix for internet services provider (ISP), the majority of planning required for their IPv6 addressing scheme will focus on the 16 bits of the subnet ID. Each ISP prefix obtain a /32 assign 65,536 sites, every customer site is given a /48 prefix from the /32 to further assign 65,536, similarly /64 prefix assign to every LAN for obtain 264 per LAN interface addresses, in some scenario there are several ways to assign, such as EUI-64 auto random number, DHCPv6, or manual configuration. IPv6 address structure is different from IPv4 and therefore a service provider has to design a new addressing plan. This process should be accomplished appropriately, in order to better utilize pool of IPv6 addresses and do not waste IP addresses space. With under this consideration to the size of the IPv6 addresses space, it may cause terribly massive routing table size. If we are talking about an IP transit service provider, then most of the providers apply IPv6 aggregation prefix filter.

IPv6 Address Prefix used for study

To implement IPv6 in research work, a unicast site local IPv6 address prefix is required, whereas every global IPv6 unicast prefix has already officially set apart as reserved by the Internet Corporation for Assigned Names and Number (ICANN) and not yet IPv6 unicast prefix has been allocated for the purpose of use in research models. Further, after acknowledgment inside the Asia Pacific regional addressing community has recommended an IPv6 prefix hold for documentation usage. As conventional method of allocating IP address space, in order to documentation of IPv6 scenarios. The IPv6 prefix assigned for research work purposes is 2001:DB8::/32.

In this article, we will discuss, the address space: 2001:10::/48. Hence, we used 16 bits from lowest order of 64 bits for design IPv6 addressing Plan. In this manner we can utilize rest of IPv6 blocks from 2001:10:0000::/64 until 2001:10:ffff::/64 to make subnets, in a sum of 65536 subnets. More simplifying we can design plan for explicitly mapping with IPv4 address subnet, in our case we have IPv4 network /8 subnet mask. For instance, a IPv4 network 10.0.0.0/8, we can utilize second last number for this mapping. So the location 10.0.0.0/8 will relate to the IPv6 address 2001:10:10::/64. This technique is extremely valuable for service providers since they can easily remember and troubleshoot IPv6 address and quickly connect that address with the particular IPv4 address. Under this assumption, they will utilize the design of IPv4 addresses into IPv6 addresses for their network topology.

IPv6 Addressing design consideration

For the IP address plan, IETF recommended various RFCs, some have been revealed. Such as RFC 4380, RFC 4291, and RFC 3849, which is based on following how IP addresses are used. In this article we considered the RFC 3849, that is mostly used IPv6 address prefix reserved for documentation. Here we will discuss how a services provider might develop their IPv6 addressing plan. To assign IP network prefix based on the network topology. We plan to utilize initial 3 bits from allocated IPv6 prefix for both the APNIC region and network prefix, as shown in Table 1.

Similarly, /36 network address assigned for infrastructure network. This is appropriate for the larger ISP network. Moreover, four IPv6 /48 prefixes are assigned users in order to provide scalability option for future usage to both customers and network. Finally, a /128 prefix are allocated for loopbacks address, as described in table 2.

Table 1 IPv6 Address Plan Distribution to Network Infrastructure

Table 2 IPv6 Loopback Address for Network Infrastructure

Designing of IPv6 Prefix Scheme for Network Infrastructure

For this article, we have taken example of an ISP for implement IPv6 in their network. The design of IPv6 address plan covers at the country level network infrastructure based on /32 prefix of IPv6. Hence it is effectively planned in a real ISP network. Where each node has allocated a /36 prefix from the parent /32 prefix, which are further subdivided into two blocks. A /64 for infrastructure and /128 blocks to the customer level. As shown in figure 2.

Figure 3 IPv6 addresses assigned ISP network

SUMMARY

The purpose of this article is to describe approach used to design IPv6 address for an ISP’s network topology by implementation dual stack transition technique. ISP network topology consists of three main area and each area have considered two routers. These are named as Node-1, Node-2, Node-3, Node-4, Node-5 and Node-6 respectively. Every router is directly connected with other routers, as shown in figure 3. The designed network topology will ensure a smooth and successful implementation of IPv6 in service provider network. Since we want a transition based on dual-stack we must assure that all services and equipment in the network can communicate through IPv4 and IPv6. After all the steps presented in this article we will further implement of IPv6 in network.

References

1. https://www.infoblox.com/wp-content/uploads/2016/04/infoblox-whitepaper-ipv6-addressing-plan-basics_1.pdf
2. Narten, Thomas, Richard Draves, and Suresh Krishnan. “Privacy extensions for stateless address autoconfiguration in IPv6”. NWG, No. RFC 4941. 2007. [Online]. https://www.hjp.at/doc/rfc/rfc4941.html
3. Enache, Daniel, and Marian Alexandru. "A Study of The Technology Transition From Ipv4 To Ipv6 For An Isp." Review of the Air Force Academy vol. 14, no. 1, pp. 117–122, 2016.
4. Huston, G., A. Lord, and P. Smith. “IPv6 address prefix reserved for documentation”. Network Working Group, No. RFC 3849. 2004. [Online]. Available: https://www.rfc-editor.org/rfc/pdfrfc/rfc3849.txt.pdf.
5. https://forum.huawei.com/enterprise/en/adoption-of-ipv6-around-the-world/thread/748889-861



Hope this helps. In case of any questions, please do let me know in the comments.