Topic 3: INTRODUCTION TO COMPUTER NETWORKS

By the end of this topic, students must be able to:

1 describe various types of computer networks
2 describe different network topologies
3 describe the OSI reference model
4 describe the TCP/IP model
5 compare the OSI reference model and the TCP/IP model
6 describe network protocols at each layer of the reference models

A computer network is an interconnection of computers and other networking devices for the purpose
of exchanging information and resources.
TYPES OF NETWORKS
 There are mainly four types of networks namely:
a. Personal Area Network
b. Local Area Network
c. Metropolitan Area Network
d. Wider Area Network

A. PERSONAL AREA NETWORK

o It is a computer network organized around an individual person.
o PAN can connect a single computer with a mobile phone or a PDA.
o You can use these networks to transfer data such as music, pictures or emails.
CHARACTERISTICS OF PAN
 A PAN can be linked by cables or wireless e.g. Bluetooth or infrared.
 It generally covers a range of less than 10 metres.
B. LOCAL AREA NETWORK
o It is a network of computers covering a single building.
o For example a network within the same building or school campus.

CHARACTERISTICS OF LAN
 It is linked by either a cable or wireless (Wireless Local Area Network-WLAN)
 It is implemented within the same building e.g. same house or school campus.
 It does not need the use of public lines.

C. METROPOLITAN AREA NETWORK

o It is a network that spans a cross the city
o Thus, it links office buildings that are geographically located within the city.
o For example a company can have three branches, one in Nkhotakota BOMA, another one at
Mwansambo and the other one at Mkaika. If these branches are on one network, this
network is referred to as MAN.
CHARACTERISTICS OF MAN
 It is implemented within a city
 It may use public lines
 It can cover several Kilometers such as 50Km.

D. WIDER AREA NETWORK

o This is a network that spans country wide or worldwide.
o This can be simply referred to as INTERNET because it spans the entire global.
CHARACTERISTICS OF WAN
 It uses microwave relays and satellites to reach users over long distances.
Microwave relay is a combination of receiving, amplifying and transmitting equipment capable
for receiving, amplifying and retransmitting analog and digital signals.
 It cannot be implement without public lines
 It cannot be used without networking devices such as routers.
E. STORAGE AREA NETWORKS
o This is an emerging network that utilizes cloud computing
o Data and programs are located remotely on a network of powerful computers and simply
accessed by users.
CHARACTERISTICS OF SAN
o Involve powerful processors, huge memories and storage capacities
o Data and programs are accessed remotely.
o Access of resources is anywhere using phones, laptops etc.
COMMON TERMS USED IN COMPUTER NETWORKING
1. Node:
- It is any device that is physically connected to a network.
- it can be a computer, printer, networking device or data storage device
2. client
- It is a node that requests and uses resources available from other nodes usually from a host
computer.
- In most cases it is a user`s microcomputer.
3. Host computer
- It is any computer in the network that gives out data to other nodes.
- It is usually a large centralized computer such as a mainframe or server.
4. Network operating system
- It is a software that controls and coordinates activities between computers and other devices
on the network.
5. Protocol
- A protocol is a rule that governs communication.
- Protocol can also be defined as a language that allow nodes to communicate with each other
over network.
  Sending and receiving systems need to use the same protocol for communication to be possible
 However if different networks with different protocols are to communicate, gateways are used in
between the networks to translate one protocol to relevant protocol for the other network.
 COMMON PROTOCOLS available include;
i. Transfer Control Protocol
 TCP defines how data are transferred across the Internet to their destination.
ii. Internet Protocol (IP)
 IP defines how data are divided into chunks called packets, for transmission.
 It also determines the path for each packet takes between computers.
6. Server
- It is a computer that supplies services or data to other machines on a network.
- A server allows two or more computer users from different work stations to have access to
the same information.
TYPES OF SERVERS
a. FILE SERVER
 This server stores files and folders that are used by other machines on the network.
 It can hold applications, text documents or user`s My documents folder.
 For security, many shared folders are stored on file servers.
b. PRINT SERVER
 This is a computer that has a printer attached to it and shares the printer for use on the
network.
c. APPLICATION SERVER
 This is a computer that has installed programs that also run on other computers in the
network.
 Database servers and e-mail clients are common way of using application servers.
d. WEBSERVER
- This is the computer that hosts and manages websites for the internet or intranet.
NETWORK TOPOLOGIES
 A network topology refers to the arrangement of nodes or how data is passed from one computer to
another in the network.
 Network topologies can be viewed as logical and physical topology

A. PHYSICAL NETWORK TOPOLOGIES

- This refers to physical arrangement of nodes on the network.
TYPES OF PHYSICAL NETWORK TOPOLOGIES
There are many types, however, this book discusses the main ones and these are:
 Point-to- point topology
 Star topology
 Ring topology
 Bus topology
 Hierarchical/tree topology
 Mesh topology.
a. POINT TO POINT TOPOLOGY
- It consist of two connected computers.
- The network contains exactly two hosts connected back to back using a single piece of cable.

- Data transmission is faster

- However, it may needs a lot of cabling in large networks which will result in mesh topology.

b. STAR TOPOLOGY
- A star topology is designed with each node connected to a central hub, switch or concentrator.
- Nodes communicate across the network by passing data through the hub or switch.
- The hub or switch manages and controls all functions of the network.
- Control is maintained by polling where each device is asked (polled) whether it has message to
send. Each device is then allowed to send.

Advantages of star topology

i. It is easy to configure
ii. If one component breaks (except a central node), the other components are not affected.
iii. It is easy to detect faults and to remove parts.
iv. It allows centralization of key resources such as servers and concentrators.
Disadvantages of star topology
i. The whole network is affected if the hub, switch or concentrator fails.
ii. It is expensive because it requires more cable length and central hub which are costly.
iii. It requires more installation time because each node forms a segment on its own.

c. BUS TOPOLOGY
- All devices are connected to the central main cable called bus or backbone.
- A terminator is attached at each end to avoid signals bouncing back and forth on the cable.
- As information passes along the bus, each node checks if the information is addressed to it or not.
 Advantages of bus topology
i. Cheap to install because it requires less cable length
ii. Easy to connect a computer or peripheral to a linear bus
iii. Reliable
Disadvantages of bus topology
i. If the main cable breaks, the entire network shut down
ii. It is relatively slow
iii. It is difficult to identify the problem if the entire network shuts down.
iv. Increase in number of computers increases collision as machines compete for transmission.

d. RING TOPOLOGY
- Devices are connected to two other devices forming a ring.
- There is no central file server or computer.
- Messages are passed around until they reach the correct destination.
- A token (something as an envelope) is used to exchange data from one station to another.

Advantages of a ring topology

i. It is cheap because it requires less cable length
ii. It is supported by several hardware vendors.
iii. Ring topology is simple to install
Disadvantages of a ring topology
i. If one component breaks down, the entire network does not work.
ii. Adding or removing components from the network requires extensive rewiring
iii. Data transfer is slower than other topologies i.e. 10Mbps
iv. Troubleshooting may be difficult.

e. HIERARCHICAL/TREE TOPOLOGY
- It is a hybrid topology
- Groups of star-configured networks are connected to a linear bus backbone.
- Therefore they utilize the advantages of star and bus topologies and improve on the disadvantages
of each topology
B. LOGICAL TOPOLOGY
 Logical topology also called signal topology deals with how data passes from one device to the
other on the network.
 This means two networks using the same physical topology can have different logical topologies.
TYPES OF LOGICAL TOPOLOGIES
 These are; Ethernet topology and Token ring topology.
a. ETHERNET TOPOLOGY
- All computers listen to the network media and can only send data when none of the others is
sending.
- Sending of data uses broadcasting technology in which data is sent on the common bus and all
nodes connected to the bus can see it. However the data has an address of the recipient and only
this computer picks the message.
b. TOKEN RING TOPOLOGY
- In this topology, data is packed in a bag like object called token and goes around the network in a
unidirectional manner.
- The token can only be taken up by the computer whose address is on the data held in the token.
- After taking it, the computer reads the data and then releases the token. The token can then be
captured by another computer which needs to transmit.
REFERENCE MODELS (RM)
 Forms an abstract framework for understanding and implementing networks.
 The reference models are based on a small number of unifying concepts and may be used as a basis
for education and explaining standards to non-specialist.
TYPES OF REFERENCE MODELS
i. OSI reference model
ii. TCP/IP reference model

A. OSI REFERENCE MODEL

 Open Systems Interconnection (OSI) is a standard reference model for communication between two
end users in a network.
 OSI reference model has 7 layers.
 It defines a networking framework for implementing protocols in seven layers.
 Control is passed from one layer to next layer starting at application in one station, proceeding to the
bottom layer, over the channel to the next station and back, up the hierarchy.
 OSI reference model was proposed by the International Standards Organisation (ISO).
 The seven layers are divided into two groups.
 The upper 4 layers are used when the message passes from or to user.
 The lower 3 layers are used when a message passes through the host computer.
 Messages intended for this computer pass to the upper layers.
 Messages destined for some other host are passed up to the upper layers but are forwarded to another
host.

THE SEVEN LAYERS OF OSI REFERENCE MODEL

LAYER 7: APPLICATION LAYER

 It is the highest layer in which user applications run.
 It generates requests for data transmission.
 At a receiving device, in the application layer is where information is opened.
 At this layer, data is in inform of application protocol datagram unit. (APDU)
 Popular protocols that are used in this layer include:
 Simple Mail Transfer Protocol (SMTP)
 File Transfer Protocol (FTP)
LAYER 6: PRESENTATION LAYER
 It formats data for transmission i.e. it converts incoming and outgoing data from one format to
another.
 Encryption and decryption of data occurs here.
 Data compression and decompression also occurs here.
 Presentation layer sometimes is called syntax layer.
 At this layer data is inform of presentation protocol datagram unit (PPDU).
 At this layer, there are no well-defined, protocols, data is only converted into required file
formats such as encoding ASCII to EBCDIC or vice versa, JPEG, TIF, GIF, MPEG etc.
LAYER 5: SESSION LAYER
 Provides session establishment, maintenance and termination.
 Provides session support such as;
- Performing security
- Name recognition
- Allowing communication to occur over network.
 Ensures synchronized data exchange.
 At this layer data is inform of Session Protocol Datagram Unit (SPDU)
 Popular protocols that are used this layer include;
 Remote Procedure Call (RPC)
 Point –to- Point Tunneling Protocol (PPTP)
 OSI Session Layer Protocol (ISO-SP)
LAYER 4: TRANSPORT LAYER
 This layer manages end-to-end transmission.
 It ensures that all packets have arrived correctly and checks for errors.
 It ensures complete data transfer through:
o Message segmentation –splitting the message into smaller units.
o Message acknowledgement –it acknowledges the receipt of all data, this is done at the
receiving end layer.
o Message traffic control –tells the transmitting station to back-off when no message buffers
are available.
o Session multiplexing –multiplexes several message streams into logical link and keeps
track of which messages belong to which sessions.
 Data at this layer is inform of transport protocol datagram unit (TPDU)
 Popular protocols that are used at this layer include:
 Transport Control Protocol (TCP)
 User Datagram Protocol (UDP)
 Sequenced Packet Exchange (SPX)
LAYER 3: NETWORK LAYER
 It handles the routing of data.
 It deals with sending and receiving information over the network.
 The network layer provides:
 o Routing: it is the forwarding of data packets across the network towards their destinations
using a router.
o Congestion control: routers can instruct a sending station to throttle back its frame
transmission when router`s buffer fills up.
o Logical-address mapping: - translates logical addresses or names into physical address.
 Data at this layer is inform of packet.
 Protocols that are commonly used at this layer includes:
 Internet Protocol (IP)
 Internetwork Packet Exchange (IPX)
 Internet Control Message Protocol (ICMP)
 Open Shortest Path First (OSPF)
LAYER 2: DATA-LINK LAYER
 It packs data into frames.
 It adds error checking information.
 It sends the frames to physical network.
 Checks received information for errors.
 Frame traffic control –tells the transmitting mode to back-off when no buffers are available.
 Frame sequencing –transmits / receives frames sequentially.
 Frame acknowledgement
 Media access management –determines when the node has the right to use physical medium.
 Data at this layer is inform of frames.
 Protocols that are used at this layer include;
o Point-to-Point Protocol (PPP)
o Ethernet
o Token ring
o Wireless access protocol (WAP)

LAYER 1: PHYSICAL LAYER

 It is responsible for ultimate transmission of data over communication media.
 Data is encoded- 1s and 0s are put in the form that can be carried by the physical media from
the source to destination.
 Data is inform of bits and is transmitted as voltage levels.
 At physical layer, no protocols are used. Data is only encoded to suit the type of media
carrying the data signal.

B. TCP/IP REERENCE MODEL

 This protocol was developed to enable internetworking on the internet before the OSI
reference model.
 It combines the transport control protocol (TCP) and the internet protocol (IP) into one.
 Its layered structure is similar to OSI. The only difference is that this has got 4 layers
instead of seven.
 Each layer of the TCP/IP model corresponds to on e or more layers of the seven-layer
OSI reference model.
 THE STRUCTURE OF TCP/IP
4. The application layer
3. The transport layer
2. The internet layer/ network layer
1. The network interface/Data-link layer
NOTE: TCP/IP does not have the presentation, session and data link layers like ISO OSI model.

LAYER 4: APPLICATION LAYER

 It is the top layer of TCP/IP protocol suite.
 It includes applications or processes that use transport layer protocols to deliver the data to
destination computers.
 At each layer there are protocol options to carry out the task designated to that particular layer.
 Some of the popular application layer protocols are:
 HTTP (Hypertext Transfer Protocol)
 FTP (File Transfer Protocol)
 SMTP (Simple Mail Transfer Protocol)
 SNMP (Simple Network Management Protocol)

LAYER 3: TRANSPORT LAYER

 This layer provides backbone to data flow between two hosts.
 It receives data from the application layer.
 Commonly used protocols at this layer are:
 TCP (Transport Control Protocol)
 UDP (User Datagram Protocol)

i. TCP (Transport Control Protocol)
 This protocol segments the data from the upper layer (Application) into chunks and passes
them to network layer.
 It acknowledges the receipts files and waits for acknowledgement of the files it sent and sets
timeout to resend the packets if the acknowledgements are not received in time.
ii. UDP (User Datagram Protocol)
 This protocol provides a simpler but not reliable service by sending packets from one host to
another.
 UDP does not care whether the sent data has been received or not.

LAYER 2: INTERNET/NETWORK LAYER

 The main purpose of this layer is to handle the movement of data on the network.
 Movement of data means routing of data over network
 The main protocols that are used at this layer are:
 IP (Internet Protocol)
 ICMP (Internet Control Message Protocol)
 IGMP (Internet Group Management Protocol)
LAYER 1: NETWORK INTERFACE/ DATA-LINK LAYER
 This layer consists of device drivers in the operating system and NIC attached to the system.
 These device drivers and NIC enables communication with media being used to transfer data over
network.
 Mostly, this media is in form of cables.
 Some of the famous protocols that are used at this layer include:
 ARP (Address Resolution Protocol)
 PPP (Point to Point Protocol)

COMPARISON BETWEEN OSI AND TCP/IP PROTOCOL SUITES

ISO OSI TCP/IP TCP/IP PROTOCOLS

Application layer
Presentation layer Application layer SMTP FTP TELNET
Session layer
Transport layer Transport layer TCP UDP
Network layer Internet layer IP IGMP etc.
Data link layer Network Interface Layer Ethernet, Token Ring etc.
Physical layer

DIFFERENCES BETWEEN OSI AND TCP/IP REFERENCE MODELS

OSI TCP/IP
1. Has 7 layers 1. Has 4 layers
2. Transport layer guarantees delivery of packets 2. Transport layer doesn’t guarantee delivery of
packets
3. Follows a horizontal approach 3. Follows vertical approach

4. Has separate presentation and session layers 4. Does not have separate presentation and session
layers
5. It is a general model such that it can be used in 5. Cannot be used in any other applications
other applications
6. Network layer provides both connectionless 6. Network layer provides only connectionless
and connection oriented services services.
7. Defines clearly services, interfaces and 7. Does not clearly distinguish between service,
protocols interface and protocols
8. Protocols are better hidden and can be easily 8. It is not easy to replace the protocols
replaced as technology changes
SIMILARITIES

1. They have similar architecture i.e. they both have a layered design.
2. They share a common application.
3. The network layer in OSI and the internet layer in TCP/IP perform same functions.
4. Both models support packet switching and routing on networks.