Analyzing the IPv6 Address Structure
IPv6 address is large (2128) allowing for many levels of subnetting and allocation of addresses between the Internet backbone and individual subnets in an organization. This permits allocation of not one but many unique IPv6 addresses to a network entity. Every address is used for a different purpose. The addresses that are provided by IPv6 are of two types - first, that are the same as IPv4 address types and second, that are unique to IPv6. A node can have many IPv6 addresses with each having a purpose of its own. The article discusses the syntax and different categories of IPv6 addresses.
IPv6 Address Syntax
The 128 bit IPv6 address is subdivided into blocks of 16 bit each and every block is converted to 4 digit hexadecimal numbers. The colons act as separators and the method of representation is known as colon-hexadecimal. A Global unicast IPv6 address is the same as a IPv4 public unicast address. Removing the preceding zeros allows simplification of the address in a block of 16 bits. It is compulsory that every block have a digit. An adjacent series of sixteen bit blocks that set to 0 in colon hexadecimal format can be compressed to ::.
IPv6 Address Prefixes
The prefix of the address indicates the bits with fixed values or bits that help identify the network. IPv6 prefixes are also represented using the slash as in the case of IPv4 address. Multinetting is a technique that allows assignment of multiple subnet IDs on the same link.
IPv6 Address Types
IPv6 addresses are of three types. They are discussed in detail in the following section.
- Unicast: This type of address is used for identifying a single interface that exists in the scope of unicast address type. Only a single interface receives a packets with a unicast address. IPv6 supports the following addresses (unicast) - Special, Global, Link-local, Site-local, Network Service Access Point (NSAP) and Internetwork Packet Exchange (IPX)addresses and Global unicast addresses.
- Global Unicast Addresses: The IPv6 of this type are the same as in the case of IPv4 public addresses. The can be routed globally and can be reached on the IPv6 internet. The addresses can be combined to produce a routing infrastructure that is efficient. It is also known as aggregatable global unicast address. The address is unique all throughout the IPv6 Internet. FP (Format Prefix) of a global unicast address is contained in the three most significant bits that are always 001. The next 13 bits are known as the Top Level Aggregator and are allocated by the IANA. The Next Level Aggregator (NLA) is contained in the next 24 bits. The Site Level Aggregator (SLA) is contained in the next 16 bits which are used for organizing addressing and routing for downstream ISPs and fo identification of sites and subnets. The next 64 bits represent the interface within a subnet and is termed as Extended Unique Identifier (EUI-64).
- Link-Local Addresses: These are the same as IPv4 addresses autoconfigured via the APIPA. These use the prefix 169.254.0.0/16. The address can be identified by the FP 1111 1110 10 followed with zeros that are 54 in number. The addresses are used nodes while communicating with nodes in the neighborhood on the same link.
- Site-Local Addresses: Though these are not preferred but a substitute for the same has not been agreed upon. These are widely used in enterprise environments. These are similar to the IPv4 private address space. The addresses can be allocated by using address configuration that is stateful as in the case of DHCPv6. A stateful address configuration is used by a host on receipt of advertisement messages by the router. Address configuration can also use a combination of stateless and stateful configuration.
- Special Addresses: There exist two special IPv6 addresses - first, the unspecified address and the loopback address. The unspecified address is used to point at the absence of an address and is the same as an IPv4 unspecified address 0.0.0.0.
- NSAP and IPX Addresses: These are used for identifying labels for network endpoints that are brought into use in Open Systems Interconenction (OSI) networking. These are also used for connecting to an Asynchronous Transfer Mode (ATM) network. IPX is not used any longer.
- Multicast: Multiple interfaces can be identified using a multicast address. This type of address identifies multiple interfaces. All interfaces that are identified by the address receive the packets with a multicast address. A multicast address enables a packet to be sent to multiple hosts that are identified by the same address. It has an FP of 11111111.
The flag settings are held in the flag field and the scope of IPv6 is in the scope field. Routers bring into use multicast scope alongwith the information that is provided by multicast routing protocols for determining if multicast traffic can be forwarded. The group ID is representative of the multicast group and is unique in a scope. While permanent assigned group IDs are scope independent, transient group IDs are applicable only to limited scope.
- Anycast: These type of addresses identify multiple interfaces. All interfaces that are the nearest and identified by the address receive the packets with a anycast address. It is important that the routing infrastructure know the interfaces with anycast addresses and the distance with reference to routing metrics. Presently, these are used as destination addresses only and are assigned to routers alone.