2-4

Chapter 2

Planning a TCP/IP Network Infrastructure

IP Addresses and Subnet Masks

IP addresses are typically expressed using dotted decimal notation, in which an

address consists of four integers—often called quads, octets, or bytes—between 0

and 255, separated by periods. Like an IP address, a subnet mask consists of 32 bits.

In decimal form, the subnet mask appears much like an IP address. In binary form,

each of the 32 bits has a value of 0 or 1. When you compare a subnet mask with an

IP address, the address bits that correspond to the 1 bits in the mask are the network

identifier bits. The address bits that correspond with the 0 bits in the mask are the

host identifier bits. For example, a typical IP address and subnet mask, expressed in

the decimal notation used when configuring a TCP/IP computer, appears as follows:

IP address: 192.168.32.114

Subnet mask: 255.255.255.0

When you convert the address and mask into binary notation, they appear as follows:

IP address: 11000000 10101000 00100000 01110010

Subnet mask: 11111111 11111111 11111111 00000000

Because the first 24 bits in the subnet mask have the value 1, this indicates that the

first 24 bits in the IP address make up the network identifier. The final eight bits in the

mask have the value 0, which means that the final eight bits in the address are the

host identifier. If the subnet mask value were 255.255.0.0 instead, this would indicate

that the network identifier and host identifier each consists of 16 bits. The division

between the 1 and 0 bits can occur almost anywhere in the subnet mask, as long as

both the network and host identifiers are each at least two bits long.

Using Registered Addresses

To prevent IP address duplication on the Internet, an administrative body called the IANA

functions as the official IP address registrar. To connect computers directly to the Internet,

you must obtain a network address from the IANA. A network address is just a network

identifier. The administrators of the network using that identifier are responsible for assign-

ing unique host identifiers to the individual computers and other devices on the network.

By combining the network identifier assigned by the IANA with a unique host identifier, the

administrators are able to calculate the IP addresses for the computers on that network.

Off the Record

Although the IANA ultimately assigns all Internet network addresses, network

administrators today do not deal with the address registrar directly. Instead, they obtain a network

address from an Internet service provider (ISP). The ISP might have obtained the network address

from a local (LIR), national (NIR), or regional Internet registry (RIR) (which is assigned pools of

addresses by the IANA directly), but it is also likely that the ISP obtained the address from its own

service provider. Internet addresses often pass through several layers of service providers in this

way before they get to the organization that actually uses them.