See also:  DNS Topics

A crash course in TCP/IP

At first glance, TCP/IP (Transmission Control Protocol/Internet Protocol) may seem baffling. Many other protocols, such as NetBEUI and IPX/SPX, require no configuration. TCP/IP is different. Due to the seemingly endless number of options that you can configure within TCP/IP, many people become intimidated at first. In reality, however, TCP/IP isnít very difficult, but you have to gain some understanding of what youíre configuring. So, Iím going to give you a crash course in TCP/IP. Although I wonít be able to explore every feature in detail, Iíll cover the important points.

The IP address

The most basic element of TCP/IP is the IP address. The IP address is a number thatís unique to each computer. If you know a computerís IP address, you can communicate with that computer from anywhere in the world. Since TCP/IP is the protocol that the Internet uses and since Internet servers are located all over the world, TCP/IP must be routable. Thus, when you try to access an IP address, your computer must be able to tell whether or not that IP address is located on your local network. If the desired address is located on your local network, you wonít have a problem reaching it. If it isnít on your local network, TCP/IP must know which network the IP address is located on in order to reach the address.

The network number represents the network that contains a given IP address. If you look through the various tabs of the TCP/IP properties sheet, youíll see that thereís no field that allows you to specify the network number. Instead, the network number is part of the IP address.

An IP address is composed of a network number and a computer number. Your computer can distinguish those two numbers because of something called the subnet mask. The subnet mask is located in a field directly below the IP address on the TCP/IP properties sheet. A simple subnet mask would be something like The numbers that make up the subnet mask indicate which portion of the IP address is the network number and which portion is the computer number. The four numbers of the subnet mask correspond directly to the four numbers in the IP address. For example, if you had a computer with an IP address of and a subnet mask of, the first two numbers in the subnet mask (both are 255) indicate that the first two numbers of the IP address are the network number. The second two numbers (both are 0) indicate that the second two numbers of the IP address are the computer number. Therefore, in the IP address, the 147.100 portion denotes which network the computer is located on, and the 100.25 portion represents a particular computer on that network. Of course, subnet masks become much more complex than this example. For instance, you can subdivide an individual network.

The default gateway

Now that you know how TCP/IP determines whether a destination address is located on a local network or on a foreign network, you may wonder how it attaches to a foreign network, especially if that network is halfway around the world. Well, it completes this job by using routing tables. Routing tables, which are stored in your routers, tell the router where to connect in certain situations.

You should note that one of the TCP/IP configuration options is for a default gateway. The default gateway is the address where a TCP/IP packet is sent if TCP/IP canít locate the destination IP address on the local network. Usually, the default gateway points to the router that controls your buildingís link to the outside world (often through an Internet connection). Since this router has a table of other routers, it knows the address and location of these routers. The destination IP address helps the router determine to which router to pass the packet. Typically, a TCP/IP packet is passed through several routers before it arrives on the destination network and, finally, on the destination PC.


If all of this information is making your head spin, thereís one TCP/IP feature that will make your life easier: Dynamic Host Configuration Protocol (DHCP). DHCP allows you to set one or more servers on your network to act as a DHCP server. When a client computer thatís set to use the DHCP option comes online, the DHCP server automatically configures TCP/IP on this computer. Of course, you still have to configure the DHCP server, but using DHCP saves you from all of the effort of configuring each client manually. It also keeps you from accidentally using a particular IP address more than once. DHCP is also useful if you have a limited number of IP addresses to work with; only the computers that are turned on at a given moment will use IP addresses.


On Windows-based networks, each computer has a name. The WINS option can be configured to contain the IP address of a Windows NT Server thatís running a WINS service. The WINS service resolves computer names (NetBIOS names) to IP addresses. Thus, if you try to access a computer by its name, your computer will go directly to the WINS database and search for the name and IP address. It wonít have to query every computer on the network to determine the right IP address. As you can imagine, using WINS really cuts down on network traffic.


DNS functions similarly to WINS, except that it manages domain names rather than computer names. A domain name is the type of name that you use on the Internet. For example, is a domain name. When you point your browser to, your computer will query a DNS server for the IP address that corresponds to the domain name. If the DNS server doesnít know the address, it asks another DNS server. This process continues until a DNS server somewhere along the line knows the address. Then, that server informs your computeróand all of the other DNS servers that are involvedóof the domain nameís IP address.


TCP/IP is more intimidating than other protocols because there are so many different options to configure. Iíve provided you with a brief explanation of how TCP/IP works. Now, you should have a better understanding of those options.

Thanks to