IPv6 Addressing

IPv6 addresses are 128 bits long. IPv6 addresses are identifiers for individual interfaces and for sets of interfaces. You can assign IPv6 addresses of all types to interfaces, rather than hosts and routers as in IPv4. Because each interface belongs to a single node, any of the interface's unicast addresses can be used as an identifier for the node. A single interface can be assigned multiple IPv6 addresses of any type.

IPv6 addresses consist of the following types: unicast, anycast, and multicast.

• Unicast addresses identify a single interface.

• Anycast addresses identify a set of interfaces. A packet that is sent to an anycast address is delivered to a member of the set.

• Multicast addresses identify a group of interfaces. A packet that is sent to a multicast address is delivered to all of the interfaces in the group.

In IPv6, multicast addresses replace broadcast addresses.

IPv6 supports addresses that are four times the number of bits as IPv4 addresses, that is, 128 in contrast to 32. Thus, the number of potential addresses is four billion times the size of the IPv4 address space. Realistically, the assignment and routing of addresses requires the creation of hierarchies that reduce the efficiency of address space usage. Consequently, the reduction in efficiency reduces the number of available addresses. Nonetheless, IPv6 provides enough address space for the foreseeable future.

The leading bits in the address specify the type of IPv6 address. The variable-length field that contains the leading bits is called the format prefix (FP). The following table shows the initial allocation of these prefixes.

Table 1-1 Format Prefix Allocations

The allocations support the direct allocation of aggregate global unicast addresses, local-use addresses, and multicast addresses. Space is reserved for Network Service Access Point (NSAP) addresses, Internetwork Packet Exchange Protocol (IPX) addresses, and neutral-interconnect addresses. The remainder of the address space is unassigned for future use. The remaining address space can be used for expansion of existing use. For example, the space can be used for additional aggregate global unicast addresses. The remaining space can also be used for new uses. For example, the space can be used for separate locators and separate identifiers. Notice that anycast addresses are not shown here because anycast addresses are allocated out of the unicast address space.

Approximately 15 percent of the address space is initially allocated. The remaining 85 percent is reserved for future use.

Unicast Addresses

IPv6 unicast address assignment consists of the following forms:

• Aggregate global unicast address

• Neutral-interconnect unicast address

• NSAP address

• IPX hierarchical address

• Site-local-use address

• Link-local-use address

• IPv4-capable host address

Additional address types can be defined in the future.

Aggregate Global Unicast Addresses

Aggregate global unicast addresses are used for global communication. These addresses are similar in function to IPv4 addresses under classless interdomain routing (CIDR). The following table shows their format.

Table 1-2 Aggregate Global Unicast Addresses Format

The first 48 bits represent the public topology. The next 16 bits represent the site topology.

The first 3 bits identify the address as an aggregate global unicast address. The next field, TLA ID, is the top level in the routing hierarchy. The next 8 bits are reserved for future use. The NLA ID field is used by organizations that are assigned a TLA ID to create an addressing hierarchy and to identify sites.

The SLA ID field is used by an individual organization to create its own local addressing hierarchy and to identify subnets. Use of the SLA ID field is analogous to subnets in IPv4 except that each organization has a much greater number of subnets. The 16-bit SLA ID field supports 65,535 individual subnets. The Interface ID is used to identify interfaces on a link. The Interface ID must be unique on that link. The Interface ID can also be unique over a broader scope. In many instances, an interface identifier is the same. In many instances, an interface identifier is based on the interface's link-layer address.

Local-Use Addresses

A local-use address is a unicast address that has only local routability scope. A local-use address can only be used within the subnet or within a subscriber network. These addresses are intended for use inside of a site for plug and play local communication. These addresses are also used for bootstrap operations for the use of global addresses.

The two types of local-use unicast addresses are link-local and site-local. The Link-Local-Use address is for a single data link layer medium, such as an Ethernet or Frame Relay link. The Site-Local-Use is for use on a single site. The following table shows the Link-Local-Use address format.

Table 1-3 Link-Local-Use Addresses Format

Link-Local-Use addresses are used for addressing on a single link for purposes such as auto-address configuration.

The following table shows the Site-Local-Use address format.

Table 1-4 Site-Local-Use Addresses

For both types of local-use addresses, the Interface ID is an identifier that must be unique in its domain. In most instances, the identifier uses a node's IEEE-802 48-bit address. The Subnet ID identifies a specific subnet in a site. The Subnet ID and the interface ID form a local-use address. Consequently, a large private internet can be constructed without any other address allocation.

Organizations that are not yet connected to the global Internet can use local-use addresses. Local-use addresses enable organizations to operate without the need to request an address prefix from the global Internet address space. If the organization later connects to the Internet, the Subnet ID, Interface ID, and a global prefix can be used to create a global address. For example, the organization can use the Registry ID, Provider ID, and the Subscriber ID to create a global address. This enhancement is a significant improvement over IPv4. IPv4 requires sites that use private (non-global) IPv4 addresses to manually renumber when sites connect to the Internet. IPv6 automatically does the renumbering.