CHAPTER THREE

NETWORK TYPES
After defining networks in the previous section and discussing their
physical structures, we need to discuss different types of networks we
encounter in the world today. The criteria of distinguishing one type of
network from another is difficult and sometimes confusing. We use a few
criteria such as size, geographical coverage, and ownership to make this
distinction. After discussing two types of networks, LANs and WANs, we
define switching, which is used to connect networks to form an
internetwork (a network of networks).

3.1 LANS, WANS, and internetworking

Local Area Network
A local area network (LAN) is usually privately owned and connects some hosts in a

single office, building, or campus. Depending on the needs of an

organization, a LAN can be as simple as two PCs and a printer in
someone’s home office, or it can extend throughout a company and
include audio and video devices. Each host in a LAN has an identifier, an
address, that uniquely defines the host in the LAN. A packet sent by a host
to another host carries both the source host’s and the destination host’s
addresses.
In the past, all hosts in a network were connected through a common
cable, which meant that a packet sent from one host to another was
received by all hosts. The intended recipient kept the packet; the others
dropped the packet. Today, most LANs use a smart connecting switch,
which is able to recognize the destination address of the packet and guide
the packet to its destination without sending it to all other hosts. The
switch alleviates the traffic in the LAN and allows more than one pair to
communicate with each other at the same time if there is no common
source and destination among them. Note that the above definition of a
LAN does not define the minimum or maximum number of hosts in a LAN.
Below figure shows a LAN using either a common cable or a switch.

Compiled By Dagne G. Page 1

Wide Area Network
A wide area network (WAN) is also an interconnection of devices
capable of communication. However, there are some differences between
a LAN and a WAN. A LAN is normally limited in size, spanning an office, a
building, or a campus; a WAN has a wider geographical span, spanning a
town, a state, a country, or even the world. A LAN interconnects hosts; a
WAN interconnects connecting devices such as switches, routers, or
modems. A LAN is normally privately owned by the organization that uses
it; a WAN is normally created and run by communication companies and

leased by an organization that uses it. We see two distinct examples of

WANs today: point-to-point WANs and switched WANs.
A point-to-point WAN is a network that connects two communicating

devices through a transmission media (cable or air). A switched WAN is

a network with more than two ends. A switched WAN, as we will see
shortly, is used in the backbone of global communication today. We can

say that a switched WAN is a combination of several point-to-point WANs

that are connected by switches.

Internetwork
Today, it is very rare to see a LAN or a WAN in isolation; they are
connected to one another. When two or more networks are connected,
they make an internetwork, or internet. As an example, assume that
an organization has two offices, one on the east coast and the other on

Compiled By Dagne G. Page 2

the west coast. Each office has a LAN that allows all employees in the
office to communicate with each other. To make the communication
between employees at different offices possible, the management leases
a point-to-point dedicated WAN from a service provider, such as a
telephone company, and connects the two LANs. Now the company has an
internetwork, or a private internet.

3.2 Peer to peer Vs Server-based Networks

Another method used to classify networks focuses on the roles the
networked computers play in the network’s operation, and more
specifically on which computer controls that operation. There are two
basic types of role classifications for networks—peer-to-peer networks and
server-based networks. The difference between the two revolves around
which computer is in charge of the network.

Peer-to-Peer Networks
A peer is considered an equal. All computers on a peer-to-peer network can be considered
equals. That is to say, no one computer is in charge of the network’s operation. Each
computer controls its own information and is capable of functioning as either a client or a
server depending on which is needed at the time.
Peer-to-peer networks are popular as home networks and for use in small companies
because they are inexpensive and easy to install. Most operating systems (the software that
runs the basic computer functionality) come with peer-to-peer networking capability built in.
The only other cost involved with setting up a peer-to-peer network comes into play if a
computer does not have a network interface card, or NIC (the device that physically connects
your computer to your network’s cabling), already installed.
Typical initial peer-to-peer networking involves no security measures. Rather, each peer
simply shares its resources and allows others open access to them. In fact, a peer-to-peer
network becomes difficult to manage when more and more security is added to the resources.

Compiled By Dagne G. Page 3

This is because users control their own security by adding password protection to each share
they create. Shares are any resources users control on their computers, such as document
folders, printers, and other peripherals. Each shared resource can actually have its own
password. Someone wanting access to numerous shared resources has to remember many
passwords. Security on a peer-to-peer network can quickly become complex and confusing.
While peer-to-peer networks are inexpensive to set up, they are extremely limited in scope.
The accepted maximum number of peers that can operate on a peer-to-peer network is ten.
They are, therefore, not appropriate for larger, more secure networks.

Server-Based Networks
Unlike peer-to-peer networks that operate without central control and are difficult to secure, a
server-based network offers centralized control and is designed for secured operations.
While there are still both clients and servers on a server-based network, a dedicated server
controls the network. A dedicated server is one that, for all practical purposes, operates
solely as a server.
A dedicated server on a server-based network services its network clients by storing data,
applications, and other resources, and then providing access to those resources when called
for by a client. When a client requests a resource such as a document, the server sends the
whole resource (the document) over the network to the client, where it is processed and later
returned to the server for continued storage. Dedicated servers can also control the entire
network’s security from one central location or share that control with other specially
configured servers.

3.3 switching
An internet is a switched network in which a switch connects at least
two links together. A switch needs to forward data from a network to

Compiled By Dagne G. Page 4

another network when required. The two most common types of switched
networks are circuit-switched and packet-switched networks.

Circuit-Switched Network
In a circuit-switched network, a dedicated connection, called a circuit, is always available
between the two end systems; the switch can only make it active or inactive. Below figure
shows a very simple switched network that connects four telephones to each end. We have
used telephone sets instead of computers as an end system because circuit switching was very
common in telephone networks in the past, although part of the telephone network today is a
packet-switched network.
The four telephones at each side are connected to a switch. The switch
connects a telephone set at one side to a telephone set at the other side.
The thick line connecting two switches is a high-capacity communication
line that can handle four voice communications at the same time; the
capacity can be shared between all pairs of telephone sets. The switches
used in this example have forwarding tasks but no storing capability.
Let us look at two cases. In the first case, all telephone sets are busy; four
people at one site are talking with four people at the other site; the
capacity of the thick line is fully used. In the second case, only one
telephone set at one side is connected to a telephone set at the other
side; only one-fourth of the capacity of the thick line is used. This means
that a circuit-switched network is efficient only when it is working at its full
capacity; most of the time, it is inefficient because it is working at partial
capacity. The reason that we need to make the capacity of the thick line
four times the capacity of each voice line is that we do not want
communication to fail when all telephone sets at one side want to be
connected with all telephone sets at the other side.

Compiled By Dagne G. Page 5

Packet-Switched Network
In a computer network, the communication between the two ends is done in blocks of data
called packets. In other words, instead of the continuous communication we see between two
telephone sets when they are being used, we see the exchange of individual data packets
between the two computers. This allows us to make the switches function for both storing and
forwarding because a packet is an independent entity that can be stored and sent later. Below
figure shows a small packet-switched network that connects four computers at one site to four
computers at the other site.
A router in a packet-switched network has a queue that can store and forward the packet.
Now assume that the capacity of the thick line is only twice the capacity of the data line
connecting the computers to the routers. If only two computers (one at each site) need to
communicate with each other, there is no waiting for the packets. However, if packets arrive
at one router when the thick line is already working at its full capacity, the packets should be
stored and forwarded in the order they arrived. The two simple examples show that a packet-
switched network is more efficient than a circuit switched network, but the packets may
encounter some delays.

3.4 Network cabling and Topologies

Network Cabling
In other words, network cabling is the medium through which information usually moves
from one network device to another. There are several types of cable which are commonly
used with LANs. In some cases, a network will utilize only one type of cable, other networks
will use a variety of cable types. The type of cable chosen for a network is related to the
network’s topology, protocol, and size. Understanding the characteristics of different types of
cable and how they relate to other aspects of a network is necessary for the development of a
successful network.

Compiled By Dagne G. Page 6

Laying cables is the foundation for both creating local area network (LANs) and connecting
LANs into wide area networks (WANs). Network administrator are usually involved only in
the planning and laying of LAN cabling since WAN cabling is the responsibility of
telecommunications carriers.

Overview of Network Topologies

The term topology refers to the way a network is laid out, either physically or logically. A
topology can be considered as the network’s shape. There are four basic topologies: Bus,
Ring, Star, and Mesh.

Bus Topology
Bus topology uses a common bus or backbone ( a single cable) to connect all devices with
terminators at both ends. The backbone acts as a shared communication medium and each
node (file server, Workstations, and peripherals) is attached to it with an interface connector.
As the message passes each station, the station checks the message’s destination address. If
the address in the message matches the station’s address deliver it.

Advantages of Bus Topology

 Connecting a computer or peripheral to a linear bus is easy.

 This topology requires least amount of cabling to connect the computers and
therefore, less expensive than other cabling arrangement.
 It is easy to extend a bus since two cables can be joined into longer cable with a
connector

Disadvantages of Bus Topology

 Entire network shuts down if there is a failure on the backbone

 Heavy traffic can slow down a bus because computers on such networks do not
coordinate with each other to reserve time to transmit.

Compiled By Dagne G. Page 7

Ring Topology
Computers are placed on a circle of cable without any terminated ends since there are no
unconnected ends. Every node has exactly two neighbors for communication purposes. All
messages travel through a ring in the same direction (clockwise or counterclockwise) until it
reaches its destination.

Advantage of Ring Topology

 Ring topology is easy to install and reconfigure.

 Every computer is given equal access to the ring. Hence, no single computer can
monopolize the network.

Disadvantages of Ring Topology

 Failure in any cable or node breaks the loop and can take down the entire network.
 Maximum ring length and number of nodes are limited.

Compiled By Dagne G. Page 8

Star Topology
Star topologies are not directly linked to each other but they are connected via a centralized
network component known as hub or concentrator. The hub acts as a central controller and if
a node wants to send data to another node, it boosts up the message and sends the message to
the intended node. Mainly used guided media

Advantages of Star Topology

 Star topology is easy to install and wire.

 The network is not disrupted even if a node fails or is removed from the network.
 Fault detection and removal of faulty parts is easier in star topology.

Disadvantages of Star Topology

 It requires a longer length of cable.

 If the hub fails, nodes attached to it are disabled.
 The cost of the hub makes the network expensive as compared to bus and ring
topology.

Compiled By Dagne G. Page 9

Mesh Topology
In mesh topology every node has a dedicated point-to-point link to every other node.
Messages sent on a mesh network can take any of several possible paths from source to
destination. A fully connected mesh network has n(n-1)/2 physical links to link n devices.
Every device on the network must have n-1 communication (input/output) ports.

Advantages of Mesh Topology

 The use of large number of links eliminates network congestion.

 If one link becomes unusable, it does not disable the entire system.

Disadvantages of Mesh Topology

 The amount of required cabling is very large.

 As every node is connected to the other, installation and reconfiguration is very
difficult.
 The amount of hardware required in this type of topology can make it expensive to
implement.

Compiled By Dagne G. Page 10