Networking 101

Status
Not open for further replies.

BuckingFastard

Baseband Member
Messages
73
The following is a crash course in basic networking and is meant for people with a little more computer knowledge than the average joe, although I'm sure the average joe could understand all this with some effort.

Protocol Introduction

Networking and the Internet is comprised of things called protocols. Protocols are like languages: if someone is speaking English and another is speaking French, they can�t communicate. The same thing applies to protocols; if two computers are using two different protocols, they can�t communicate.

TCP/IP

The most prominent protocol is TCP/IP (Transfer Control Protocol / Internet Protocol). TCP/IP guarantees the delivery of information. In other words, if information is being sent from one computer to another, if the information gets lost along the way then the computer will resend that information until it arrives at its destination.

TCP/IP is comprised of four main parts: IP address, MAC address, subnet mask, and default gateway. An IP address is a 32-bit number that represents a logical identification to a computer. Since the address is logical, it doesn�t actually exist somewhere in the computer�s hardware. It can be changed and is controlled by software. Generally an IP address can range from 1-254 and is broken into four 8-bit segments. An IP address appears in the format �xxx.xxx.xxx.xxx�. An example of an IP address could be �68.55.210.74�.

A MAC address is something that every network interface card (NIC). Every MAC address is unique and is burned into the card by the manufacturer. A MAC address cannot be changed. It is 192-bit number that represents the physical identification of a computer and is displayed in hexidecimal (http://en.wikipedia.org/wiki/Hexadecimal). It is split into 12 segments of 16 bits and appears in the format XX-XX-XX-XX-XX-XX. An example of a MAC address could be 00-05-5D-6F-F5-5F. The first three pairs represent the manufacturer of the NIC card (such as 3com or Linksys) and the last three are used only for that manufacturer. In other words, the last three pairs uniquely identify a NIC card to a specific manufacturer (represented by the first three pairs).

A subnet mask is a little more complicated and will be covered later. For now, just know that two computers usually have to have the same subnet mask to talk to each other. The subnet mask is also a 32-bit number and appears in the same format as an IP address. Usually you will see subnet masks contain either 255 or 0, such as �255.255.255.0�.

A default gateway is where the computer will send all its information if it doesn�t know where else to send it. For instance if I want to send information to a certain computer, my computer will check to see if it knows where the destination computer is. If it doesn�t, it sends the information to the default gateway hoping that it will know what to do with it. A default gateway is an IP address of some networking device (computer, router, etc).

Other Protocols

Some other popular protocols include IPX/SPX which is used for Novell (an operating system like Windows or Mac OS), AppleTalk (used for Mac OS), and NetBEUI (used for Windows in some cases).


Networking Media

There are different ways to connect devices together and each has its advantages and disadvantages. A few examples of different networking media are Ethernet, Fiber, Coax, and Wireless.

Ethernet is the normal networking cable that you can buy in your local computer shop. It has a connector that looks similar to a phone jack connector; however it�s a little bigger. This connector is called an RJ-45 connector. Ethernet uses copper and pulses of electricity to send information.

Fiber is used for running networks across long distances, such as across country or even through a building. Fiber uses glass and pulses of light to send information.

Coax is usually no longer used; however it was in the 70�s, and 80�s. Coax is the same cable that you hook up to your TV or to your cable modem. It uses copper and pulses of electricity to send information.

Wireless is growing in popularity and comes in three flavors: A, B, and G. There are advantages and disadvantages to each. Each of the three uses RF (radio frequency) to send information.

802.11A runs in the 5GHz range and has a theoretical maximum speed of 54Mb/s (Mega bits per second (There are eight Mega bits (Mb) in a Mega Byte (MB), so 54Mb/s is equal to 6.75MB/s)). Running at 5GHz may be good because microwaves and 2.4GHz wireless phones cannot interfere with your wireless signal. However due to the high frequency at which it runs, it does not have the range of 802.11B or 802.11G.

802.11B and 802.11G both run in the 2.4GHz range and are both susceptible to 2.4GHz phones and microwaves which can interfere with the signal. Because of the lower frequency, both B and G have a higher range than A, somewhere around 100 feet. 802.11B has a theoretical maximum of 11Mb/s (1.375 MB/s) and 802.11G�s theoretical maximum is 54Mb/s (6.75MB/s).

Using proprietary hardware, speeds of over 100Mb/s can be attained using wireless.

Also, please note that, unless you are a big business, your internet connection will always be slower than wireless. Your wireless connection will never be slower than your Internet connection. Period.


Networking Hardware

There are five basic types of networking hardware: routers, hubs, switches, NICs, and transceivers.

Hubs

A hub is a simple device that passes along information. When computers are connected to a hub, they can talk to each other. The advantages of a hub are that it�s very inexpensive. The disadvantages of a hub are that it doesn�t look at the information being passed through to determine where it goes. It just copies the information and sends it to everything it is connected to.

So if I have 300 computers connected with hubs, when one computer sends out a piece of information all 299 computers will receive it. Every computer except for the one the information is destined for will simply ignore it. This isn�t a problem when you have a small amount of computers, however when you have around 50 or more computers you may notice a slowdown.

Switches

You can think of a switch as a smart hub. The advantage of a switch is that it passes information to computers, however it knows what is connected to it. Although this is not entirely correct, for the purposes of this tutorial we can say that a switch will only send information to the computer is it destined for. It will not send the information to everything connected to it. A disadvantage to a switch is that they are usually more expensive than hubs.

Routers

Routers may seem complicated, but essentially a router is just a traffic cop. It knows what networks it has connected to it, and when it receives information it sends it to the correct network. A network can simplistically be defined as a group of computers with the same logical addresses (for example IP addresses).

NICs

A NIC (network interface card) is something that enables a computer to be connected to a network. A NIC can be used for all kinds of networking media (Ethernet, Wireless, etc).

Transceivers

A transceiver is something that connects two different types of networking media. For example, a wireless access point is a transceiver; it converts Ethernet to a wireless signal.

Well, that's Networking 101. Stay tuned for the next installment: Networking 102! Comments/criticisims are welcome!
 
That's nice of you to provide some info to help people get a basic grasp of networking. :)

I have to disagree on a few things, though.

1. TCP alone (without IP) provides reliable data transmission. IP alone is a unreliable protocol.

2. Switches aren't smart hubs. Switches could be called smart bridges, though. Hubs and switches are quite different, and not really comparable like that.
 
I agree. Things like that which are a little more in depth will be covered in later editions.

But for the average person, a switch and hub do basically the same thing, even though they operate on different layers.
 
if i made a transfer cable(switching green and orange on one side of the cable) will all that stuff other than the hardware remain the same?
 
I'm not sure I fully understand your question. The wires themselves and crimping the cable will remain the same and be the same process.

Making that type of cable is called a cross-over cable. Essentially, you need a cross-over cable when you're connecting two similar devices (switch -> switch, router -> router, PC -> PC).
 
BuckingFastard said:
Essentially, you need a cross-over cable when you're connecting two similar devices (switch -> switch, router -> router, PC -> PC).

True, but not so much anymore. I have a cross-over cable handy sometimes but, honestly, the last time I've used one was sometime (years) ago in the cisco academy.
Now, and for years, simple hubs have had an uplink port or a port that can switch from either/or. Switches have dedicated uplink ports. Smart switches or managed switches have up and down ports for just that purpose. And even the NICs that I've been putting into PCs have been auto-sensing in the past year or two in normal production. I've had more than 100 situations where a cross-over cable would've "normally" come in handy....but a straight-through was just fine. All hardware is different, but now-a-days, 99 times out of 100, you don't need 'em.
 
Status
Not open for further replies.
Back
Top Bottom