How the Internet Works

What is a network?
What is TCP/IP
Packet Switching
The Domain Name System
Domains

What is a Network?

A LAN or Local Area Network is essentially two or more computers directly linked together (usually by cable) in a room or a building. This type of connection enables users to share common resources, such as printers or a Fileserver (a central computer that stores information and software).

Each computer connected to a LAN can share information and communicate with every other computer and peripheral connected to the LAN.

When two or more LAN’s are connected, you create a WAN (Wide Area Network). The LAN’s that comprise the WAN may be on different floors of the same building, in different buildings or in different countries. Most WAN’s are connected by dedicated, high speed telephone lines. Sophisticated networks may even use satellites to share or swap data between the LANs which comprise the WAN.

In essence, the Internet is a massive Wide Area network connecting hundreds of smaller computer networks around the world.

Back to TOP

What is TCP/IP

All computers connected to the Internet share a common language, or more accurately a ‘protocol’, known as TCP/IP.

TCP/IP is an acronym for Transmission Control Protocol/Internet protocol. This is a standard that regulates how all computers connected to the Internet must act when communicating with one another.

Every computer connected to a network running TCP/IP software is aware of every other computer on that network, thereby knowing the exact position of each computer on the ‘network map’

Knowing this, a computer can send information to the intended recipient by the fastest route without having to follow a predetermined path (which might otherwise have been blocked or congested). If it finds that a computer is not responding or is too busy to handle the information, then it would simply consult the network map (referred to as the network topology) to find another computer that can pass on the information.

TCP/IP is in fact two protocols in one. The TCP part of the protocol handles the way in which data is sent across a network while the IP part handles how computers locate one another.

Back to TOP

Packet Switching

When information is sent from one computer to another across a network, whether it is a short email message or a large document file, it is not all sent at once. Rather, it is broken down into smaller pieces (called ‘packets’), which are sent one at a time.

The reason files are broken into packets is fairly straightforward. To enable several computers to send information over the same telephone line, no single computer is allowed to monopolise the line. If this were the case, a computer sending a very large file would prevent others from sending information while it used the line.

To ensure all computers have equal access to the line, the information they are sending is broken into packets. These are sent down the line interspersed with packets from other computers. In this way, all computers can use a single line at the same time.

To ensure packets don’t get lost, each contains information about the Internet addresses of the intended recipient and the original sender. They are also sequentially numbered so the recipient’s computer can correctly reassemble them. The packets are passed from computer to computer until they reach their destination.

For instance, you may send email from your computer in Perth to a friend in Tokyo. While there is not a direct Internet connection from your city to Tokyo, Missourinet's computers will know, from the email address you supplied, exactly where on the Internet your friend is, and will pass the message on to a computer closer to Tokyo.

That computer might not be in Tokyo either, but it will send the packets on to a computer even closer to the destination. This ‘pass the packets’ procedure will continue until eventually all the packets are received by the Tokyo computer.

As you can see the packets comprising your email message may pass through the hands of many computers before they are received and reassembled, then placed in the recipients email box.

All the packets might not follow the same path, but may be handled by different computers. Wherever they may pass, rest assured they will all end up at the specified email address.

It should be noted that the above steps (breaking the email or file into packets, individually ordering them, passing them on, being inspected by different hosts along the way) occur in fractions of a second. It should take less than a minute for email to get from one end of Australia to the other, and only a little longer than that to get from one end of the world to the other.

Back to TOP

The Domain Name System

The Domain Name System (DNS) is the means by which the thousands of separate and diverse networks are ‘mapped’. Every time you connect to a computer over the Internet or send email to a colleague, you use the DNS. It is essentially a collection of large databases that are used by computers on the Internet to locate other Internet computers.

Every computer connected to the Internet is given an IP address. For the technically minded, an IP address is a 32-bit numerical address, represented by four 8-bit numbers which are expressed as decimal numbers in the range of 0-255 (inclusive) separated by periods ‘.’ . Each 8-bit component of the 32-bit address is referred to as an “octet”.

An IP address might look something like this: 203.59.24.221

When your computer establishes a connection to one of Missourinet’s dial-up access points, it is assigned an IP address, so that other computers connected to the Internet can transmit information to it as required.

Because humans have difficulty remembering numbers represented in such a fashion, computers are also given human-friendly names (known as domain names) such as: missourinet.net Which is significantly easier to remember.

Computers prefer to communicate with each other using numerical IP addresses. So in order for humans to use the addresses they can remember, the two names are cross-referenced.

The DNS takes care of this cross-referencing. When you specify an Internet address using the human-friendly domain name, your computer will first access a database known as a DNS nameserver, which contains both the human-friendly and numerical addresses of all computers connected to the Internet.

If it finds the domain name you specify, it will look up the corresponding numerical address, and use that to carry out the requested function. As a result, you need never bother memorising numerical IP addresses.

The DNS also imposes a uniform naming system onto an otherwise chaotic network of networks. For each of the separate networks that make up the Internet to act cohesively , they must recognise each other’s existence and be able to transmit information without any confusion as to exactly where it is going.

Back to TOP

NEXT TOPIC >>>

<<< PREVIOUS TOPIC