|
|
|
 |
 |
 |
 |
 |
 |
|
|
|
|
|

Internetworking Basics |
|
|
|
|
|
 |
 |
|
 |
| |
Cabling the LAN
Cabling the LAN
These notes guide you through the understanding of cabling LAN's.
Introduction
Ethernet Media and Connectors
Connectors
Media and their Connectors
UTP Cable Wiring
Straight-through Cable
Crossover Cable
Rollover Cable
LAN Connection Devices
Repeaters and Hubs
Bridges and Switches
Routers
Workstation and Server Roles
Peer-to-Peer Networks
Client Server Networks
Summary
| |
| |
Introduction
On completion of these notes you should...
-
know which connectors are required for each different media
-
know the difference in the wiring of straight-through, crossover and rollover cables and the devices you can connect with each cable
- understand what client and server roles are
-
understand the difference between peer-to-peer networks and client-server networks
|
| |
Ethernet Media and Connectors
The TIA/EIA-568-B standard specifies that copper cable has to pass ten tests to be acceptable for use on modern networks. These tests are the subject of these notes.
Connectors
|
Standard |
|
Topology |
Max Length |
Connector |
| 10BASE2 |
802.3a |
coaxial (thinnet) |
Bus |
185m |
BNC |
| 10BASE5 |
802.3 |
coaxial (thicket) |
Bus |
500m |
AUI |
| 10BASE-T |
802.3i |
UTP CAT 3,4,5 |
Star |
100m |
RJ45 |
| 100BASE-TX |
802.3u |
UTP CAT 5 |
Star |
100m |
RJ45 |
| 100BASE-FX |
802.3u |
62.5 micron multimode fiber |
Star |
400m |
ST or SC |
| 1000BASE-T |
802.3ab |
UTP CAT 5, 6 |
Star |
100m |
RJ45 |
| 1000BASE-SX |
802.3z |
50 or 62.5 micron multimode fiber |
Star |
275m for 62.5 micron.
550m for 50 micron |
SC |
| 1000BASE-LX |
802.3z |
50 or 62.5 micron multimode fiber or 9 micron single-mode fiber |
Star |
440m for 62.5 micron.
550m for 50 micron.
3 to 10 km for single-mode. |
SC |
Media and their Connectors
|
| |
UTP Cable Wiring
UTP cable contains eight wires, twisted together in four pairs. An RJ45 connector is crimped onto each end os the cable. Inside the connector are eight terminals that are referred to as pin 1 to pin 8. For a cable to pass the EIA/TIA standard, the cable wires must be arranged in a specific order with respect to the pins inside the connectors. There are two standards that can be followed, EIA/TIA-586-A and EIA/TIA-586-B. The first is common in America while the second is common in Europe.
When creating a cable, as well as deciding which standard to follow the type of cable also needs to be decided on. Straight-through cables or crossover cable.
Straight-through Cables
Straight-through cable is a wiring arrangement where the wires are connected to the same pin at each end. A straight-through wiring arrangement that follows the EIA/TIA-586-B is shown below.
The same wire is connected to pin 1 at both ends. Another wire is connected to pin 2 at both ends, and so on.
For CAT 5 cable, used on fast Ethernet LAN's, only wires on pins 1,2,3 and 6 are used for transmitting signals. The other wires are unused. On Gigabit LAN's, all the wires are used.
You can use straight-through cable for linking computers or routers to hubs or switches. |
|

|
Crossover Cables
Crossover cable is a wiring arrangement where the wires are not all connected to the same pin at each end. A crossover cable wiring arrangement that follows the EIA/TIA-586-B is shown below.
The wires connected to pins 1 and 3 at one end are crossed over to connect to pins 3 and 1 at the other end.
The wires connected to pins 2 and 6 at one end are crossed over to connect to pins 6 and 2 at the other end.
You can use crossover cable for linking similar devices together, such as computer to computer, router to computer, hub to hub, switch to switch or switch to hub. |
|

|
Rollover Cables
A rollover cable is different to the other two cables. It is used for connecting a PC to the console port of a router. You create this sort of link only when you wish to configure the router. A rollover cable wiring arrangement is shown below.
For a rollover cable, the wires at one end are all crossed over at the other end.
Imagine wiring the cable at one end following the pattern for a straight-through cable.
Then you wire the other end also following the same pattern for a straight-through cable. However, this time the connector is held so it faces down instead of the normal way up. |
|

|
This is where the name' rollover' comes from because the wiring arrangement looks like you have rolled the cable over before wiring the second end.
~~Activity~~
| Activity A |
Examine the picture below. Two cables, crossover and straight-through, are attached to a PC. Which devices could you connect to each cable?
- Hover your mouse over each hyperlink below the diagram to see suggestions.
- Decide if each suggestion is valid.
- Click on each hyperlink to sees the answer.

1. Is this allowed? 2. Is this allowed? 3. What about this? |
|
| |
LAN Connection Devices
A LAN is made up of many different types of devices. The devices discussed below, repeaters, hubs, bridges and switches are commonly found on networks. Switches are in widespread use and are probably the most prevalent of all.
Repeaters and Hubs
UTP cable is commonly found on networks today. However, one of the disadvantages of this cable is that the maximum length allowed is only 100m. This is because signal attenuation causes a signal to become weaker as it travels along a cable. A repeater can be used to amplify and regenerate a weakened signal.
Most networks require links between devices longer than 100m and may use repeater or hubs to extend the distance. For example, a repeater may be used to link two lengths of UTP cable together, extending the distance to 200m.
You can keep on adding cable and repeaters up to a certain overall limit. The overall distance is regulated by the 5-4-3 rule. This rule means you are allowed up to 5 network segments can be connected together, linked by 4 repeaters but only 3 segments can have hosts connected. Although it is easy to apply this rule to bus networks, it is harder when considering networks arranged in extended star topologies or when switches are used.
The reason for the overall limit is to do with collisions on CSMA\CD networks, where a device at one end of a network needs to be able to detect collisions at any point along the network. If there are too many segments and the furthest point is too far away then collisions will not be detectable.
A hub is similar to a multiport repeater. The purpose of a hub, like that of a repeater is to retime, regenerate and amplify signals. However hubs are commonly used as central concentration devices, with workstations typically arranged in a star topology. |
|
|
One of the disadvantages of hubs is the way in which they relay network traffic to all connected devices, regardless of the end destination of the traffic. This means that hubs and attached devices hear all traffic; also collisions are much more likely. Hubs therefore are said to maintain a single collision domain.
~~Activity~~
| Activity B |
Examine the picture below. Two cables, crossover and straight-through, are attached to a hub. Which devices could you connect to each cable?
- Hover your mouse over each hyperlink below the diagram to see suggestions.
- Decide if each suggestion is valid.
- Click on each hyperlink to sees the answer.

1. Is this allowed? 2. Is this allowed? 3. What about this? |
Bridges and Switches
Like hubs and repeaters, bridges and switches are also used to connect network segments. A bridge usually connects two segments together, whereas a switch normally act as a central connection device, linking many devices together.
Both bridges and switches are much more intelligent than hubs because they inspect network traffic and make decisions on where to send the traffic depending on the source and destination MAC addresses. Thus, traffic is much more regulated:-
- A bridge inspects the source and destination address inside a frame and if both are for devices on the same segment, the bridge prevents the traffic from reaching other segments. This reduces unnecessary traffic and is known as filtering. If the destination is on a different segment then it forwards the the frame to that segment. If the bridge does not know the destination address, it forwards the frame to all segments - this is know as flooding.
|
|
|
 |
Just like a bridge, it makes forwarding decisions based on the source and destination address contained in a frame. However, you can consider a switch to be even more intelligent than a bridge, since it has many ports, whereas a bridge normally just has two.
Every single port connection on a switch creates a separate segment between the port and the device attached at the other end.
|
Switches are one of the most common devices used networks today. They can act as central connection points like hubs, which they are slowly replacing because they improve performance twofold - by reducing traffic congestion and by reducing the size of collision domains.
~~Activity~~
| Activity C |
Examine the picture below.Two cables, crossover and straight-through, are attached to a switch. Which devices could you connect to each cable?
- Hover your mouse over each hyperlink below the diagram to see suggestions.
- Decide if each suggestion is valid.
- Click on each hyperlink to sees the answer.

1. Is this allowed? 2. Is this allowed? 3. What about this? |
Routers
Unlike a switch, a router is more concerned with routing network traffic along the most efficient path to the destination. The destination host can be on a different LAN segment or even a completely different network.
Routers maintain routing tables containing IP addresses. When packets arrive a router makes forwarding decisions based on its routing tables. This decision making is similar to a switch and its MAC address table.
Routers can also connect different LAN technologies together such as Ethernet, Token Ring, FDDI and ATM.
Routers are also used all over the Internet. In fact, the Internet could not exist in its current form without routers.
|
|
|
~~Activity~~
| Activity D |
Examine the picture below. Two cables, crossover and straight-through, are attached to a router. Which devices could you connect to each cable?
- Hover your mouse over each hyperlink below the diagram to see suggestions.
- Decide if each suggestion is valid.
- Click on each hyperlink to sees the answer.

1. Is this allowed? 2. Is this allowed? 3. What about this? |
|
| |
Workstation and Server Roles
One of the main reasons for networking computers together to form a LAN is so that they can share resources.One computer for example may have access to a printer, while another may have a special application installed, while another may contain files that need to be frequently updated and shared amongst users.
To accomplish this sharing, computers take on different roles. One computer may be configured so that other computers can share some of it's files and folders. The computer with the resource is called the server, while the computer that connects and gains access to a resource is called a client. The roles my be reversed at any time. For example, a computer is acting in the role of a client when it is accessing a remote resource located on another computer, but if it shares a resource with other computers, then its role switches to become that of a server.
There are two main categories of networks when considering roles. There is the peer-to-peer network and the client-server network.

Choosing between implementing a peer-to-peer or client-server network for a particular organization depends on a few factors, on the size of the network; the number of users, computers and other devices. On whether the organization intends the network to grow rapidly and on how much can be spent on its implementation and upkeep.
Peer-to-Peer Networks
A peer-to-peer network, also called a workgroup, is commonly used for home and small business networks. In this type of setup, computers communicate directly with each other and share resources.
The computers in a workgroup are called peers because they are all equal. Each user decides on the resources their workstation will share with others on the network. Resources that could be shared include printers, scanners, folders and files.
Peer-to-peer networks do not require a server to manage network resources. In general, a peer-to-peer network is appropriate for small networks where there are less than ten computers connected together. |
|
 |
The advantages and disadvantages of peer-to-peer networks are:-
Advantages
- Easy to set-up for small networks
- No need for a dedicated network administrator
- Less expensive to set up
- Can easily share files, folders, printers etc.
Disadvantages
- There is no central management point for sharing resources
- Peer-to-peer networks do not scale well
- Each user must be trained to share resources with others
- Not so secure as client-server networks
- The performance of a workstation may drop if a lot of users connect at the same time.
Client-Server Networks
A client-server network is commonly used for medium to large business networks. In this type of setup, a server manages the sharing of network resources.
The computers in client-server network rely on one or more central management points to access shared network resources such as printers, scanners, folders and files.
A client-server network is more appropriate for medium size to larger networks because delivery and access to resources are more easily managed from a central control point. Servers are designed to cope with many simultaneous connections from users that need to access resources at the same time.
|
|
 |
Security and access control is more easily managed with this type of network and is considered extremely important by most organizations.The network administrator that manages the server(s) decides on access rights for users. A user may be allowed access to some shared resources but denied access to others. Normally, a user must authenticate themselves to a server before accessing any resources. Administrators manage users by creating accounts for them.each with their own username and password. Managing user accounts from a central location is one of the advantages of this type of client-server networks.
Moreover, when network users save their data to file servers, administrators can back up the data easily and regularly to another location. Relying on a single server would mean relying on a single central connection point and the entire network would grind to a halt if it failed. Many client-server networks use more than one server as an extra precaution against a server failing.
The advantages and disadvantages of client-server networks are:-
Advantages
- More scaleable than peer-to-peer networks
- There is a central management point for sharing resources
- More easily administered due to the central management point
- More secure than peer-to-peer networks
- Data can be backed up from one central location
Disadvantages
- More complicated to setup than peer-to-peer networks
- A dedicated network administrator is needed
- More expensive to set up, e.g. server hardware is more expensive since it has to be a high performance machine.
- Requires more complicated and expensive network operating systems such as Windows 2000/XP/2003, Novell NetWare, UNIX or Linux.
|
| |
Summary
On completing these notes you should have learned the following key points:-
- BNC, RJ45, AUI, SC, ST are all connectors - each required for different media
- Straight-through cable is commonly use for linking computers to hubs or switches
- Crossover cable is commonly use for linking similar devices, such as two computers together or two switches or two routers or a computer to a router
- Rollover cable is used for linking a computer to the console of a router
- Repeaters and hubs are similar in that they link devices together
- Repeaters and hubs increase the size of collision domains
- Bridges and switches make forwarding decisions based on MAC addresses. They reduce network traffic and increase the number of collision domains.
- Routers make forwarding decisions based on IP addresses
- A peer-to-peer arrangement is suitable for home and small business networks
- A workstation in a peer-to-peer network may act in the role of a client or a server
- A client-server arrangement is suitable for medium to large networks
|
 |
|
 |
|
 |
|
 |
|
| |
|
|

|