CSC 812 Data Communication and Networks –

Lecture 1

This lecture aims to outline the development of computer networks. We

will also discuss various network topologies and elaborate on their
structural and functional features.

The Advancements of Computer Networks

The origin of computer networks goes as far back as 1969 with some key
moments outlined below;

ARPANET (1969): The U.S. Department of Defense developed the first

packet-switching network that enabled resource sharing between various
universities and research institutions.

The transition to TCP/IP (1983): This replaced NCP (Network Control

Protocol) standardizing communication across diverse networks, and
laying the groundwork for the modern internet.

In 1983, the network was then divided into military (MILNET) and civilian
networks (INTERNET) marking the first historical use of the term
“Internet.”

When ARPANET was decommissioned in 1990, this allowed for NSFNET

and commercial ISPs to develop, which expanded worldwide connectivity.

There are multiple architectures vital to understanding the various forms

of networks. Below are some key ones

The Development of Computer Networks

There are multiple architectures and protocols vital to understanding the

various forms of networks. Some vital ones will be discussed.

Network Topologies

A topology describes the layout of a computer network, either physically

or logically, and affects its cost, scalability, and fault tolerance.

1
Star Topology

In a star topology, all network devices (nodes) are connected to a central

hub or switch. This hub acts as a central point for data transmission, with
each node having a direct connection to it.

Structure: All nodes are connected to a central hub such as a switch.

Figure 1: An Example of Star Topology Layout

Advantages:

 Fault Isolation: The overall network remains intact since device

failure on one node does not cause a cascade effect within the
entire network.
 Ease of Management: Adding and removing nodes to the network
can be executed easily.

Limitations:

 High Cabling Costs: The costs rise with the amount of cabling
needed.
 Single Point of Failure: If the central hub fails, the entire network
goes down.

Bus Topology

In a bus topology, all network devices connect to a single, shared cable

called a "bus" or "backbone". This cable acts as a common communication
channel, allowing all devices to receive data signals simultaneously.

Structure: A single backbone cable is used as the connection point for all
nodes to attach to.

2
 Figure 2: An Example of Bus Topology Layout

Advantages:

 Low Cost: The expenditure for hardware as well as cabling is

significantly lower.
 Simplicity: Suitable for small networks since installation is not time-
consuming.

Limitations:

 Core Cable Vulnerability: Damaging the backbone cable means the

entire network stops functioning.
 Security Risks: The data is received by each node leading to the
exposure of private classified data.

Ring Topology

A ring topology connects networked devices in a closed loop, where each

device is connected to exactly two others, creating a circular path for data
transmission. This configuration allows data to flow in a single direction,
from node to node, until it reaches its destination.

Structure: Nodes are interconnected in a closed loop and data is

transmitted in one direction.

3
 Figure 3: An Example of Ring Topology Layout

Advantages:

 Efficient data flow: There is a smooth flow of data through the

network as traffic collisions do not occur due to the uni-directional
nature.
 Fault Tolerance: In the case of an error, automatic switching to a
backup system is available in dual-ring structures.

Limitations:

 Single link dependency: The whole system may become inoperative

due to one broken link.
 Multi-Level Problem Solving: Finding the cause of the problem is
difficult.

Mesh Topology

A mesh topology, in networking, is a network configuration where devices

(nodes) are interconnected directly to multiple other nodes, creating a
decentralized, web-like structure. This means data has multiple paths to
travel, leading to increased redundancy and resilience.

Structure: Nodes are interconnected through multiple routes.

4
 Figure 4: An Example of Mesh Topology Layout

Advantages:

 Redundant paths ensure continuity, providing high fault tolerance.

 Increased tolerances: Maintain efficacy in greater network loads.
 Scalability: Responsive to more powerful, larger, and dynamic
networks.

Limitations:

 High Cost: Exceeds expenses for wiring schematics and port

allocation.
 Difficult Maintenance: Complicated to configure and maintain.
 Complex Installation: Difficult configuration and maintenance.

Topology Comparison

Topology Best Use Case Reliability Cost

Star Small office networks regions Moderate Medium
Bus For temporary setups Low Low
Ring For industrial automation High (dual-ring) High
Mesh For critical infrastructure Very high Very high

5
Data Flow:

The three modes of data flow are:

Simplex:

Figure 5: Simplex Data Flow

Half-Duplex:

Figure 6: Half-Duplex Data Flow

Full Duplex:

Figure 7: Duplex Data Flow

6
Networking Modes:

The two modes of networking are:

Peer-to-Peer:

Figure 8: Examples of Peer-to-Peer Networking

Client-Server:

Figure 9: Examples of Client-Server Networking

7
Data Transmission:

The three modes of data transmission are: Unicast, Multicast, and

Broadcast.

Figure 10: An Illustration of Unicast, Multicast and Broadcast

8
The ISO-OSI Reference Model

The OSI model standardizes network communication into seven

abstraction layers, although a simplified five-layer version is often used
for teaching clarity.

Traditional 7-Layer OSI Model

Layer Function Key

Protocols/Techniques
Application User-network interaction (e.g., HTTP, SMTP, DNS
HTTP, FTP, DNS).
Presentation Data translation, encryption, and SSL/TLS, ASCII/Unicode
compression. conversion
Session Establishes, maintains, and NetBIOS, RPC
terminates sessions between
applications.
Transport End-to-end data delivery with TCP, UDP
error recovery and flow control.
Network Logical addressing (IP) and IP, ICMP, OSPF
routing across networks.
Data Link Framing, MAC addressing, and Ethernet, CSMA/CD, ARP
error detection (CRC).
Physical Transmits raw bitstreams over RS-232, 100BASE-T
physical media (e.g., cables, Wi-
Fi).

9
 Figure 11: Detailed Illustration of the Traditional 7-Layer OSI Model

Simplified 5-Layer Version

Layer Function Merged OSI

Layers
Application Combines OSI’s Application, Layers 5–7
Presentation, and Session layers.
Transport Unchanged: Ensures reliable data Layer 4
transfer (TCP/UDP).
Network Unchanged: Handles IP addressing and Layer 3
routing.
Data Link Manages MAC addressing and frame Layer 2
delivery.
Physical Transmits bits via electrical/optical Layer 1
signals.

Key Differences:

 The 5-layer model consolidates upper OSI layers (5–7) into a single
Application layer, prioritizing practicality over theoretical
separation.
 OSI’s Session and Presentation layers are omitted in the
simplified version, as their functions (e.g., encryption, session
management) are often handled directly by applications in real-
world implementations.

10
 Figure 12: Traditional 7-Layer OSI Model vs. 5-Layer OSI Model

TCP/IP Protocol Suite

The TCP/IP model, a four-layer framework, supports modern internet

communication and serves as the practical implementation of OSI
principles.

TCP/IP Layers
Layer Function Key OSI
Protocols Equivalent
Applicatio User-facing services (e.g., HTTP, FTP, Layers 5–7
n web browsing, email). DNS, SMTP
Transport End-to-end data delivery TCP, UDP Layer 4
with error control (TCP) or
low latency (UDP).
Internet Logical addressing IP, ICMP Layer 3
(IPv4/IPv6) and routing
across networks.
Network Physical transmission and Ethernet, Layers 1–2
Access MAC addressing. Wi-Fi

11
 Figure 13: TCP/IP Protocol Suite

TCP/IP vs. OSI

Aspect OSI Model TCP/IP Model
Layers 7 (theoretical) 4–5 (practical)
Design Focus Protocol-agnostic Real-world internet
standardization communication
Session/ Explicit layers (5–6) Integrated into
Presentation Application layer
Adoption Educational/descriptive Dominant in modern
tool networks

12
 Figure 14: Traditional 7-Layer OSI Model vs. TCP/IP Protocol Suite

Practical Relevance:

 TCP/IP streamlines OSI’s upper layers into a single Application layer,

reflecting how protocols like HTTP inherently manage encryption
(TLS) and session control (cookies).

 The Network Access layer in TCP/IP integrates the Physical and Data
Link layers in OSI, reinforcing the transmission and reception of bits
as independent of the hardware used.

OSI serves as a guide for its users, offering a comprehensive scheme for
studying layered communication, unlike TCP/IP which is more practical
and focused on the actual functioning of the internet. Students proficient
in both models can more effectively design networks and resolve issues
arising from multiple layers.

13