An in-depth piece on building, managing and maintaining the physical aspects of a network.

Irrespective of whether you are a three people office or a Fortune 100 behemoth, networks are an essential part of your organization. They are the nervous system that carries information and data connecting everybody and nowadays everything.

Despite their abundance, networks are largely mysterious, especially for business decision makers. This two part article is aimed at lifting the mist that covers networks.

This article is in two parts. The first part covers the Physical network, and the second part “intelligent wire” deals with Network Services that build the intelligence in a network. In each section, there is an emphasis on how to build different networks for SOHOs, SMEs and Large Enterprises.

The Physical Network

The Physical network is simply the wire and associated electronics used to connect computers and other devices. This layer can be as simple as connecting the right cable between two computers to a more complex system that includes Wireless Networks or WiFi.

A network is an interconnection of many parts. The main thing is a cable.

Picture 1.0 and 1.1 below – an Ethernet cable and Ethernet cable connected to a Ethernet port of a Computer

A port is receptacle into which the Jack goes. The name of the port depends on the technology used. Therefore, you have LAN Ports like Ethernet ports and Communication Ports like USB ports, etc.

To connect two computers, both computers need to be equipped with a similar port. By default, this means an Ethernet port or a USB Port today. A simple thing to remember with networks is to have the same accessories, technologies and software at both ends of the connection. Though this is a simple concept, you will be surprised to know the number of mistakes made breaking this rule.

To illustrate, if you have an Ethernet LAN Port on your computer at one end, you should have an Ethernet LAN port at the other computer also. Connecting an Ethernet port to a USB port will not work!

To proceed, you need to take a cable called an Ethernet cable, with Ethernet jacks at both the ends and connect each jack to the Ethernet port of each computer. Voila! You are done setting up the physical network! Let me introduce another term here; the Ethernet cable you need to use in this case is called a Cross-Over Ethernet Cable. This is a cable used to connect two computers with out any electronics in between. The section below – Addressing – mentions what needs to be done on the computers to get them to communicate or send and receive data from each other.

Picture 2 - Below – Two Computers connected by an Ethernet Wire

This is good if you want to connect just two computers. If you need to
connect more than two computers, you need a piece of network electronics hardware called a Hub or a Switch.

A Hub or a Switch is a like a telephone exchange, that connects two computers that want to communicate with each other. On average, a Switch is ten to hundred times faster than a Hub. This may not sound like much, but if you copy, say, a large photo album from one computer to another, you could save a minute to an hour in the copying process. This depends on the file size being copied. So larger the file, the more time saved by the switch. In normal applications a switch will help you backup your data faster.

Picture 3 - Hub / Switch

Picture 4 below - PCs connected to Hub / Switch

Wireless – WiFi

Suppose, some of your computers have a WiFi card installed on them, you could connect them to your network by adding a WiFi Access Point to your network. In this case, you will not need any wires and you could move from say your cabin to the conference room and take your laptop with you, and not loose the network connection.

Picture 5 below - Network with WiFi Access Point

Remove all the computers then you have a picture of a Physical Network. The non electricity powered elements like wires and jacks are called passive components and powered elements like a Switch or a Hub, or WiFi Access Point are called active components.

Structured Cabling

Setting up a small network is not very messy. Just lots of wires all over the place. However, in larger networks, I would say anything beyond 10 computers; it makes sense to structure the cabling. This simply means to use standard cable laying techniques, using structured cable components, and cable management components. These techniques help in reducing the time spent in troubleshooting or maintaining your physical network. A brief diagram of a structured physical network is provided below.

Picture 6 – Below - Components of a structured cabling system

Addressing

Finally, after connecting the computers together, they have to be able to communicate. To do that, you have to give each computer a different address. In addition to a Computer name, you will have to set an IP address on each computer. For the sake of simplicity, I normally suggest people use 10.10.10.X IP addresses. This means, the first three 10s are static, and you can keep changing the X – 1,2,3,4,…255. This configuration can connect upto 255 computers or devices to your network. You can connect more computers, by changing the third 10 in the series.

Best Practices

Connect your Active components to a UPS. They do not draw much in power and will ensure your network is still running during a power failure.

Anil Kumar is the Founder & CEO of XNAPSYS, a Leading Network Service  Appliance provider. Feedback can be e-mailed to This e-mail address is being protected from spam bots, you need JavaScript enabled to view it

Issue BG56 Nov05

Related Items:

---

Be first to comment this article

Only registered users can write comments.
Please login or register.