Understanding IP Addressing

This topic provides useful background information for someone intending to configure a networked application using TCP/IP.

Each connection point (node) on an IP network has a unique IP address. For example, a PC with two network cards would have two connection points and hence two IP addresses.

IP addresses can either be static, allocated by a configuration tool on the equipment in which the node resides, or dynamic and allocated by a network administration tool each time the node connects to the network. For a PC running Microsoft Windows, static addresses are entered as part of Windows network configuration. If you are using a dynamic address then it is allocated by a device on the network known as DHCP (Dynamic Host Configuration Protocol) server. For use with PcVue’s networking, it is recommended that you use static addressing.

The IP address structure

IP currently uses a 32 bit address allowing for a theoretical maximum of 232 addresses (approximately 4.3 billion). The address is represented as four bytes written as four decimal values separated by full stops, for example 192.168.2.2. The address is divided into two parts, the network number (often known as the subnet) and the host (or machine) number. All hosts on the same IP network must share the same network number but have a unique host number. The address split can be done in one of three ways resulting in what is known as the A, B and C address classes.

Class	Mask	1st byte	2nd byte	3rd byte	4th byte
A	255.0.0.0	Subnet number 1 to 127	Host number 0 to 255	Host number 0 to 255	Host number 0 to 255
B	255.255.0.0	Subnet number 128 to 191	Subnet number 0 to 255	Host number 0 to 255	Host number 0 to 255
C	255.255.255.0	Subnet number 192 to 223	Subnet number 0 to 255	Subnet number 0 to 255	Host number 0 to 255

Allocating IP addresses

Blocks of IP addresses are allocated by an international organization known as the NIC. This ensures that each node has a unique address worldwide thereby eliminating problems that would occur with two identical addresses on the same network. All class A addresses and most class B addresses are already allocated.

The exception to this is a range of IP addresses known as Private Addresses. These can be used for any nodes that are NOT exposed to the outside world, for example the Local Area Network (LAN) in a factory. There are three private address ranges, one in each of the address classes. For small LANs of under 255 machines the Class C range is generally used. The table below illustrates the three private address ranges.

Class	Start of range	End of range	Mask
A	10.0.0.0	10.255.255.255	255.0.0.0
B	172.16.0.0	171.31.255.255	255.255.0.0
C	192.168.0.0	192.168.255.255	255.355.255.0

Examples of class C IP addressing

Computer	IP address	Subnet mask	Network number	Machine number
Pippin	192.168.2.3	255.255.255.0	192.168.2	3
Frodo	192.168.2.4	255.255.255.0	192.168.2	4
Gandalf	192.168.22.3	255.255.255.0	192.168.22	3
Aragorn	192.168.22.4	255.255.255.0	192.168.22	4

In the above example, Pippin and Frodo are on one subnet whilst Gandalf and Aragorn are on another. Therefore, Pippin and Frodo will be able to exchange information with one another but not with Gandalf and Aragorn.

Important note about using more than one node on a computer running Windows

If you use more than one node on a computer running Windows, for example to make use of PcVue’s dual network capability, then each node must match a different subnet.