network
Worked for 9s
ICT 1216Y(1) — INTRODUCTION TO COMPUTER NETWORKS
DETAILED STUDY NOTES — LECTURES 1–4
---
LECTURE 1 — INTRODUCTION TO COMPUTER NETWORKS
1. COMPUTER NETWORK
A computer network is a group of computers and other devices that are connected to each other.
The main purpose of a computer network is to allow devices to communicate and share resources.
Examples of resources that can be shared
- Files
- Printers
- Storage
- Software
- Hardware
- Internet connections
- Network services
A network allows computers to communicate efficiently instead of operating as completely isolated systems.
---
2. HOW COMPUTERS COMMUNICATE
Communication between computers can be viewed at three high-level levels:
1. Hardware level
2. Operating system level
3. Application level
2.1 Hardware Level
At the hardware level, devices need a physical or wireless connection.
Connections can use:
- Network cables
- Wireless communication
A computer generally requires a Network Interface Card (NIC) to connect to a network.
NIC
A Network Interface Card provides the hardware interface between a computer and the network.
Two common types are:
- Wired NIC
- Wireless NIC
A wired NIC uses a physical cable, while a wireless NIC uses radio communication and an antenna.
Each NIC has a hardware address called a MAC address.
A MAC address is a 48-bit, or 6-byte, address used to identify a device at the local network level.
---
3. MAC ADDRESS
A MAC address (Media Access Control address) is a hardware-level address associated with a network interface.
Main purpose
It identifies a network interface/device on the local network.
Important characteristics
- 48 bits
- 6 bytes
- Associated with the NIC
- Used at the lower network layers
- Primarily relevant to local network communication
MAC vs IP
MAC Address| IP Address
Hardware address| Logical/network address
48 bits in the lecture notes| IPv4 = 32 bits; IPv6 = 128 bits
Identifies a device/interface locally| Used for identifying/locating devices across networks
Associated with NIC| Assigned/configured as a network address
The important distinction is:
MAC → local/hardware identification
IP → network-level identification
---
4. IP ADDRESS
An IP address is used at the network/operating-system level to identify a computer on a network.
The lecture introduces:
- IPv4 → 32-bit address
- IPv6 → 128-bit address
IP addresses are used to identify and locate devices across interconnected networks.
---
5. PORT NUMBER
At the application level, computers often run many network applications simultaneously.
A port number identifies the particular application/service that should receive network traffic.
For example, a computer can have:
- A web application
- An email application
- Other network services
The IP address identifies the computer, while the port helps identify the application/service on that computer.
Important relationship
MAC address → network interface/device locally
IP address → host/network location
Port number → application/service
This distinction is important when explaining how data reaches the correct application.
---
6. NETWORK INTERFACE CARD (NIC)
A NIC allows a computer or device to communicate over a network.
Wired NIC
Uses a physical cable.
Wireless NIC
Uses radio communication and an antenna.
The NIC contains or uses the hardware addressing required for local network communication.
---
7. HUB
A hub is a networking device used to connect multiple devices within a network segment.
When a hub receives data, it broadcasts the data to all of its ports.
Therefore, every connected device may receive the transmitted signal even when the data is intended for only one device.
Characteristics of a hub
- Broadcasts incoming data
- Sends data to all ports
- Uses shared communication
- Operates in half-duplex
- Can experience collisions
- Less efficient
- Less secure than a switch
Advantages
- Simple
- Relatively inexpensive
- Easy to use for basic connections
Disadvantages
- Unnecessary traffic is sent to all devices
- More collisions can occur
- Lower performance
- Lower security
- Shared bandwidth
Memory aid
HUB = sends to everyone
---
8. SWITCH
A switch is a networking device that connects devices and forwards data toward the intended destination.
Unlike a hub, a switch does not normally broadcast every frame to every connected device. It can forward traffic to the appropriate destination port.
Characteristics
- Sends data toward the intended destination
- Uses unicast communication where appropriate
- Can operate in full-duplex
- Reduces unnecessary traffic
- Reduces collisions
- Improves network performance
- Provides better security than a hub
Hub vs Switch
Hub| Switch
Broadcasts data| Forwards data to intended destination
Sends to all ports| Sends to appropriate port
Half-duplex| Can operate full-duplex
More collisions| Fewer collisions
Lower performance| Better performance
Lower security| Better security
Memory aid
HUB → everyone
SWITCH → intended destination
---
9. ROUTER
A router connects different networks.
For example, a router can connect:
Local Area Network → Internet
A router is therefore different from a switch.
Switch
Connects devices within a network.
Router
Connects different networks.
Memory aid
Switch = inside a network
Router = between networks
---
10. NETWORK CABLES AND TRANSMISSION MEDIA
The lecture identifies three important wired media:
1. Twisted pair
2. Coaxial cable
3. Fibre optic
10.1 Twisted Pair
Twisted-pair cable consists of pairs of copper wires twisted together.
It is commonly used in Ethernet networks.
The lecture notes that twisted-pair Ethernet can support high transmission speeds.
---
10.2 Coaxial Cable
Coaxial cable contains conductors arranged concentrically.
It resembles the type of cable traditionally used for television connections.
It can carry network signals and can support bidirectional communication.
---
10.3 Fibre Optic
Fibre optic cable transmits data using light.
Advantages
- Very high transmission speed
- Long-distance communication
- Low error rate
- Resistant to electromagnetic interference
Disadvantage
- Generally more expensive than common copper media
---
11. FEATURES OF COMPUTER NETWORKS
Important reasons for using computer networks include:
1. Communication speed
2. File sharing
3. Backup and rollback
4. Hardware sharing
5. Software sharing
6. Security
7. Scalability
---
11.1 Communication Speed
Networks allow information to be transferred rapidly.
Examples include:
- Video conferencing
- File transfers
- Online communication
---
11.2 File Sharing
Users can share files between computers instead of keeping completely separate copies.
---
11.3 Backup and Rollback
Centralized network resources can make backup easier.
If important data is stored centrally, it can be backed up and restored more efficiently.
---
11.4 Hardware Sharing
Multiple users can share hardware such as:
- Printers
- Storage devices
- Other network resources
---
11.5 Software Sharing
Network environments can allow software/resources to be managed centrally.
---
11.6 Security
Networks can control access to resources.
A network operating system can check whether a user has permission to access a particular resource.
---
11.7 Scalability
Scalability refers to the ability to add components to a network as requirements increase.
A scalable network can grow by adding:
- Users
- Computers
- Devices
- Network resources
---
12. TRANSMISSION MODES
Transmission mode describes the direction in which data can travel between communicating devices.
There are three major transmission modes:
1. Simplex
2. Half-duplex
3. Full-duplex
---
13. SIMPLEX
In simplex communication, data travels in only one direction.
One device only sends while the other only receives.
Example
Radio broadcasting:
Radio station → receivers
Advantages
- Entire communication channel can be used for transmission in one direction.
- Simple communication arrangement.
Disadvantage
- There is no communication in the reverse direction.
Diagram
Sender → Receiver
Memory aid
Simplex = one-way
---
14. HALF-DUPLEX
In half-duplex communication, both devices can transmit and receive, but they cannot do so at the same time.
Example
Walkie-talkie.
One person speaks while the other waits, then they switch.
Advantages
- Both devices can send and receive.
- The communication channel can be used in either direction.
Disadvantage
- Devices have to wait for their turn.
- Communication can therefore experience delays.
Diagram
Device A ↔ Device B
but only one direction at a time.
Memory aid
Half-duplex = both directions, one at a time
---
15. FULL-DUPLEX
In full-duplex communication, both devices can transmit and receive simultaneously.
Example
Telephone communication.
Both people can speak and hear at the same time.
Advantage
- Simultaneous two-way communication
- More efficient for interactive communication
Disadvantage
- More complex than simplex or half-duplex communication.
Memory aid
Full-duplex = both directions simultaneously
---
16. TRANSMISSION MODE COMPARISON
Feature| Simplex| Half-Duplex| Full-Duplex
Direction| One direction| Both directions| Both directions
Simultaneous?| No| No| Yes
Example| Radio| Walkie-talkie| Telephone
Reverse communication| No| Yes| Yes
---
17. COMPUTER NETWORK ARCHITECTURE
Network architecture describes how computers, services, hardware and communication are organized.
Two important architectures are:
1. Client/server
2. Peer-to-peer
---
18. CLIENT/SERVER ARCHITECTURE
In a client/server network, a dedicated server provides services or resources to clients.
Client
A client requests and uses services.
Examples:
- Web browser
- Email client
Server
A server provides services/resources.
It can manage:
- Files
- Directories
- Printers
- Security
- Network resources
---
Advantages of Client/Server
Centralized management
Resources can be controlled from a central server.
Security
Access rights can be managed centrally.
Backup
Important data can be stored and backed up centrally.
Data consistency
Central management can help maintain consistent data.
Performance
A dedicated server can be designed to provide services to many clients.
---
Disadvantages
Cost
Server hardware and software can be expensive.
Server failure
If the central server fails, services depending on it may become unavailable.
Bottleneck
A server can become overloaded if too many clients request resources simultaneously.
Maintenance
Servers require administration and maintenance.
Scalability cost
Increasing capacity may require additional server resources.
---
19. PEER-TO-PEER NETWORK
A Peer-to-Peer (P2P) network has no dedicated server.
Each peer can act as both:
- Client
- Server
The lecture describes P2P as useful for smaller environments, usually up to around 10 computers.
---
Advantages of P2P
Cost-effective
A dedicated server is not required.
Resource sharing
Files, storage and bandwidth can be shared directly between peers.
Decentralization
There is no central server controlling everything.
Robustness
Failure of one peer does not necessarily bring down the entire network.
Reduced bottlenecks
There is no single central server handling all requests.
---
Disadvantages of P2P
Security
Security is harder to manage because every peer can introduce risks.
Administration
Decentralized management makes administration more difficult.
Data consistency
Maintaining consistent copies of data across peers can be difficult.
Performance
Different peers may have different capabilities.
Scalability
As more peers are added, network traffic and coordination overhead can increase.
---
20. CLIENT/SERVER VS P2P
Client/Server| Peer-to-Peer
Dedicated server| No dedicated server
Centralized| Decentralized
Central management| Distributed management
Generally easier to administer| More difficult to administer
Server provides resources| Peers provide/share resources
More expensive| Generally cheaper
Server can be bottleneck| No central server bottleneck
Suitable for larger managed environments| Useful for smaller environments
---
21. TYPES OF COMPUTER NETWORKS
Networks can be classified according to geographical size.
The four main categories are:
1. PAN
2. LAN
3. MAN
4. WAN
Size order
PAN → LAN → MAN → WAN
---
22. PAN — PERSONAL AREA NETWORK
A Personal Area Network (PAN) is a very small network centered around an individual.
Typical range is a few metres.
The lecture notes describe approximately 2–10 devices and a typical range around 10 metres.
Examples
- Smartphone
- Laptop
- Smartwatch
- Bluetooth devices
- Wearable devices
- Tablet
Example
A smartphone connected to a smartwatch and wireless headphones.
---
23. LAN — LOCAL AREA NETWORK
A Local Area Network (LAN) covers a limited geographical area.
Examples:
- Home
- Office
- School building
- University building
- Campus
LANs generally provide high-speed communication.
Typical technologies include:
- Ethernet
- Wi-Fi
Characteristics
- Limited geographical area
- High data transfer rates
- Resource sharing
- Usually privately managed
---
24. MAN — METROPOLITAN AREA NETWORK
A Metropolitan Area Network (MAN) covers a larger geographical area than a LAN but smaller than a WAN.
It can connect multiple LANs across:
- A city
- Metropolitan area
- Large campus
The lecture notes give a typical coverage of up to approximately 50 km.
---
25. WAN — WIDE AREA NETWORK
A Wide Area Network (WAN) covers a large geographical area.
It can connect networks across:
- Cities
- Countries
- Continents
The Internet is the largest example of an interconnected WAN-like system.
---
26. NETWORK TOPOLOGIES
A network topology describes the arrangement of devices and connections in a network.
The lecture covers:
1. Point-to-point
2. Bus
3. Ring
4. Star
5. Tree
---
26.1 Point-to-Point Topology
A direct connection exists between two devices.
Device A ↔ Device B
It is simple and suitable when only two devices need a direct connection.
---
26.2 Bus Topology
All devices connect to a common backbone cable.
Components
- Backbone cable
- Drop lines
- Connectors
- Terminators
Data travels along the common communication medium.
Disadvantages
Failure of the backbone can affect the network.
---
26.3 Ring Topology
Each device is connected to two neighbouring devices.
The connections form a ring.
Device → Device → Device → Device → back to first device
Data can travel around the ring.
---
26.4 Star Topology
All devices connect to a central networking device, usually a switch.
Example:
Computer → Switch ← Computer
Advantages
- Easy to manage
- Easy to add/remove devices
- Failure of one cable generally affects only one device
- Centralized connection point
Disadvantage
Failure of the central device can affect connected devices.
---
26.5 Tree Topology
Tree topology is a hierarchical arrangement.
Multiple star networks can be connected together to form branches.
It is useful for larger hierarchical network structures.
---
27. NETWORK SECURITY
Network security protects:
- Data
- Devices
- Network resources
- Services
- Users
The three fundamental principles are:
Confidentiality
Integrity
Availability
Together they are known as the CIA triad.
---
28. CONFIDENTIALITY
Confidentiality means information should only be accessible to authorized individuals.
Example
Sensitive information should not be accessible to unauthorized users.
Common protection
Encryption.
---
29. INTEGRITY
Integrity means data should remain:
- Accurate
- Complete
- Unaltered
during storage or transmission.
Techniques such as hashing and digital signatures can help protect integrity.
---
30. AVAILABILITY
Availability means network resources and services should remain accessible when required.
A denial-of-service attack can affect availability by overwhelming a network or server.
---
31. COMMON NETWORK SECURITY THREATS
Malware
Malicious software such as:
- Viruses
- Worms
- Trojans
can damage systems or steal information.
Phishing
Fraudulent communication designed to trick users into providing sensitive information.
Man-in-the-Middle Attack
An attacker intercepts communication between two parties and may attempt to obtain or modify information.
Denial-of-Service Attack
An attacker overwhelms a network or server to disrupt its services.
---
32. NETWORK SECURITY MEASURES
Firewall
A firewall monitors and controls incoming and outgoing traffic according to security rules.
It acts as a barrier between trusted and untrusted networks.
IDS
Intrusion Detection System
Monitors network activity for suspicious behaviour.
IPS
Intrusion Prevention System
Detects suspicious activity and takes action to prevent threats.
VPN
A Virtual Private Network provides encrypted communication over a network and can provide secure remote access.
Antivirus/Anti-malware
Detects and removes malicious software.
---
33. NETWORK SECURITY BEST PRACTICES
Important practices include:
- Regular updates and patches
- Strong password policies
- Multi-factor authentication
- Network segmentation
- Security awareness training
Network segmentation divides a network into smaller sections, which can help limit access and contain security breaches.
---
LECTURE 1 — KEY EXAM POINTS
- Definition and purpose of a computer network
- NIC
- MAC address
- IP address
- Port number
- Hub
- Switch
- Router
- Network media
- Simplex
- Half-duplex
- Full-duplex
- Client/server
- P2P
- PAN, LAN, MAN, WAN
- Network topologies
- CIA triad
- Common network attacks
- Security measures
---
LECTURE 1 — FLASHCARDS
Q: What is a computer network?
A: A group of computers/devices connected together to communicate and share resources.
Q: What is the main purpose of a computer network?
A: Resource sharing and communication.
Q: What does NIC stand for?
A: Network Interface Card.
Q: What is a MAC address?
A: A 48-bit hardware address associated with a network interface.
Q: What is an IP address?
A: A logical network address used to identify/locate a host across networks.
Q: What does a port number identify?
A: An application or service on a host.
Q: What does a hub do?
A: Broadcasts incoming data to all ports.
Q: What does a switch do?
A: Forwards data toward the intended destination port.
Q: What does a router do?
A: Connects different networks.
Q: What is simplex transmission?
A: One-way communication.
Q: What is half-duplex transmission?
A: Two-way communication, but only one direction can transmit at a time.
Q: What is full-duplex transmission?
A: Two-way communication occurring simultaneously.
Q: What is client/server architecture?
A: An architecture in which clients request services/resources from a server.
Q: What is P2P architecture?
A: An architecture with no dedicated server where peers can act as both clients and servers.
Q: What are the four network size categories?
A: PAN, LAN, MAN and WAN.
Q: What does PAN stand for?
A: Personal Area Network.
Q: What does LAN stand for?
A: Local Area Network.
Q: What does MAN stand for?
A: Metropolitan Area Network.
Q: What does WAN stand for?
A: Wide Area Network.
Q: What are the three principles of the CIA triad?
A: Confidentiality, Integrity and Availability.
---
LECTURE 2 — COMPUTER NETWORK MODELS
1. LAYERED ARCHITECTURE
Network communication is complex because many different tasks have to be performed.
Originally, communication software could be designed as one large, complex and unstructured program.
This made it difficult to:
- Understand
- Test
- Modify
- Maintain
- Replace parts of the system
A layered architecture divides the networking process into separate layers.
Each layer has a specific responsibility.
---
2. PURPOSE OF LAYERED ARCHITECTURE
Divide and conquer
A complex networking task is divided into smaller and more manageable tasks.
This reduces design complexity.
Modularity
Each layer can be designed and understood independently.
Easy modification
Changes to one layer should not require major changes to other layers.
Easy testing
Individual layers can be analyzed and tested separately.
Independence
Implementation details in one layer can be hidden from other layers.
---
3. BASIC ELEMENTS OF LAYERED ARCHITECTURE
Three important concepts are:
1. Service
2. Protocol
3. Interface
---
3.1 SERVICE
A service is the set of functions/actions that one layer provides to the layer above it.
Simple definition
Service = what a layer provides
---
3.2 PROTOCOL
A protocol is a set of rules governing communication between peer entities.
A protocol defines:
- Message format
- Message order
- Actions taken when messages are sent
- Actions taken when messages are received
Simple definition
Protocol = rules of communication
---
3.3 INTERFACE
An interface defines how information is transferred between layers.
Simple definition
Interface = how layers communicate
---
4. DATA MOVEMENT THROUGH LAYERS
When a sender transmits data, information moves down through the layers.
Each layer can add information, such as a header.
At the receiving side, the data moves upward.
The corresponding layer processes/removes the relevant information.
General process
Application data
↓
Headers are added
↓
Lower layers process the data
↓
Physical transmission
↓
Receiver processes the information
↓
Headers are processed/removed
↓
Application receives the data
This process is an example of encapsulation and decapsulation.
---
5. LAYERED COMMUNICATION EXAMPLE
A simple chat application illustrates the idea.
Application layer
The user selects a message.
Presentation layer
The message can be encrypted by the sender and decrypted by the receiver.
Transport layer
A port number identifies the application.
Network layer
Source and destination IP addresses identify the communicating hosts.
Physical layer
The data is converted into signals for transmission.
The lecture's classroom activity represents binary values using hand positions to demonstrate physical transmission.
---
6. OSI MODEL
OSI = Open System Interconnection
The OSI model is a reference model describing how information from an application on one computer moves through a network to an application on another computer.
It was developed by the International Organization for Standardization (ISO) in 1984.
The OSI model contains seven layers.
---
7. SEVEN OSI LAYERS
From bottom to top:
1. Physical
2. Data Link
3. Network
4. Transport
5. Session
6. Presentation
7. Application
Mnemonic
Please Do Not Throw Sausage Pizza Away
Physical → Data Link → Network → Transport → Session → Presentation → Application
---
8. PHYSICAL LAYER
The Physical layer is Layer 1.
Its main function is to transmit individual bits from one node to another.
Main functions
Line configuration
Defines how devices are physically connected.
Transmission mode
Defines whether communication is:
- Simplex
- Half-duplex
- Full-duplex
Signals
Determines the type of signal used to transmit information.
Key word
Physical = bits and signals
---
9. DATA LINK LAYER
The Data Link layer is Layer 2.
It organizes raw bits into structured units called frames.
Main functions
Framing
Converts the raw bit stream into frames.
Headers and trailers are added.
Physical addressing
Uses physical/hardware addressing to help deliver frames to the correct device.
Error control
The lecture discusses the use of CRC for detecting errors.
If an error is detected, retransmission may be requested.
Access control
Determines which device gets access to a shared communication channel.
Key words
Data Link = frames + MAC/physical addressing + error control + access control
---
10. NETWORK LAYER
The Network layer is Layer 3.
Main functions
Internetworking
Provides logical connections between different networks/devices.
Logical addressing
Adds source and destination network addresses.
Routing
Determines an appropriate/optimal path from source to destination.
Packetizing
Converts data received from an upper layer into packets.
Key words
Network = IP + routing + packets
---
11. TRANSPORT LAYER
The Transport layer is Layer 4.
It provides end-to-end communication between applications.
Main functions
Service-point addressing
Uses port numbers to identify applications.
Segmentation
Breaks data into smaller segments.
Reassembly
Reconstructs the original data at the destination.
Connection control
Can provide:
- Connection-oriented communication
- Connectionless communication
Flow control
Controls the amount of data sent between communicating endpoints.
Key words
Transport = ports + segments + end-to-end communication
---
12. SESSION LAYER
The Session layer is Layer 5.
It manages communication sessions between applications.
Functions
- Session establishment
- Session management
- Session termination
- Synchronization
- Checkpoints
- Recovery
Checkpoints can allow communication to resume from a known point after an error.
Key word
Session = manages conversations/sessions
---
13. PRESENTATION LAYER
The Presentation layer is Layer 6.
It deals with how data is represented.
Main functions
- Translation
- Encryption
- Compression
Translation
Converts data between different encoding/representation formats.
Encryption
Protects information by converting it into a form that unauthorized users cannot easily understand.
Compression
Reduces the amount of data required for transmission/storage.
Key words
Presentation = translation + encryption + compression
---
14. APPLICATION LAYER
The Application layer is Layer 7.
It provides network-related services to applications used by the user.
Examples include applications such as:
- Web browsers
- Email applications
A text editor that does not communicate over the network is not itself a network application.
Key word
Application = network services used by applications
---
15. OSI MODEL SUMMARY
Layer| Name| Main Functions
7| Application| Network applications/services
6| Presentation| Translation, encryption, compression
5| Session| Session establishment, management, termination
4| Transport| Ports, segmentation, reassembly, flow/control
3| Network| IP addressing, routing, packets
2| Data Link| Frames, physical addressing, error/access control
1| Physical| Bits, signals, physical transmission
---
16. TCP/IP MODEL
The TCP/IP model is the model associated with Internet networking.
The lecture presents four main layers:
1. Application
2. Transport
3. Internet/Network
4. Network Interface
From bottom to top:
Network Interface → Internet → Transport → Application
---
17. TCP/IP APPLICATION LAYER
The Application layer provides network services to applications.
Examples of application protocols:
- HTTP
- DNS
- DHCP
- SMTP
---
18. TCP/IP TRANSPORT LAYER
The two main transport protocols discussed are:
- TCP
- UDP
---
19. TCP
TCP = Transmission Control Protocol
TCP provides reliable, connection-oriented communication.
Main characteristics
- Connection-oriented
- Reliable
- Uses sequence numbers
- Uses acknowledgements
- Detects errors
- Retransmits lost/damaged data
- Reorders segments at the receiver
The sender divides a message into segments.
Each segment is assigned a sequence number.
The receiver uses sequence numbers to reconstruct the original message.
---
20. UDP
UDP = User Datagram Protocol
UDP provides:
- Connectionless communication
- End-to-end delivery
- Unreliable transmission
UDP does not provide the same reliability mechanisms as TCP.
---
21. TCP VS UDP
TCP| UDP
Connection-oriented| Connectionless
Reliable| Unreliable
Uses sequence numbers| No equivalent TCP reliability mechanism
Uses acknowledgements| No TCP-style acknowledgements
Retransmits lost/damaged data| Does not provide TCP-style retransmission
Reorders segments| Does not provide TCP-style reordering
---
22. INTERNET/NETWORK LAYER
The Internet/Network layer handles IP-based communication.
The lecture associates this layer with:
- IP
- ARP
- ICMP
---
23. IP
IP provides network-level communication.
It is responsible for logical addressing and forwarding packets between networks.
---
24. ARP
ARP = Address Resolution Protocol
ARP is used to determine a physical address associated with an IP address.
Basic process
1. A device needs the physical address associated with an IP address.
2. It sends an ARP request.
3. The request is broadcast.
4. The device with the relevant IP address responds with an ARP reply.
5. The physical address can then be used for local communication.
Key idea
ARP = IP address → physical/MAC address
---
25. ICMP
ICMP = Internet Control Message Protocol
ICMP is used to communicate information about problems associated with IP datagrams/network delivery.
Key idea
ICMP = network error/control information
---
26. NETWORK INTERFACE LAYER
The Network Interface layer combines functions associated with the physical and data-link aspects of communication.
It deals with:
- Physical transmission
- Frames
- Physical addressing
- Access to the communication medium
---
27. OSI VS TCP/IP
OSI| TCP/IP
7 layers| 4 layers in the lecture
Physical| Network Interface
Data Link| Network Interface
Network| Internet/Network
Transport| Transport
Session| Application
Presentation| Application
Application| Application
The TCP/IP model combines some functions that are separate in OSI.
---
LECTURE 2 — KEY EXAM POINTS
- Purpose of layered architecture
- Advantages of layering
- Service, protocol and interface
- Data encapsulation/decapsulation
- Seven OSI layers
- Function of every OSI layer
- TCP/IP layers
- OSI vs TCP/IP
- TCP vs UDP
- IP
- ARP
- ICMP
- HTTP
- DNS
- DHCP
- SMTP
---
LECTURE 2 — FLASHCARDS
Q: Why is layered architecture used?
A: To divide complex networking tasks into smaller, manageable and independent tasks.
Q: What are the advantages of layered architecture?
A: Reduced complexity, modularity, easier modification and easier testing.
Q: What is a service?
A: Functions provided by one layer to the layer above it.
Q: What is a protocol?
A: Rules governing communication between peer entities.
Q: What is an interface?
A: The mechanism through which information moves between layers.
Q: How many layers does OSI have?
A: Seven.
Q: What is OSI Layer 1?
A: Physical.
Q: What is OSI Layer 2?
A: Data Link.
Q: What is OSI Layer 3?
A: Network.
Q: What is OSI Layer 4?
A: Transport.
Q: What is OSI Layer 5?
A: Session.
Q: What is OSI Layer 6?
A: Presentation.
Q: What is OSI Layer 7?
A: Application.
Q: Which OSI layer handles bits and signals?
A: Physical layer.
Q: Which layer handles frames?
A: Data Link layer.
Q: Which layer handles routing?
A: Network layer.
Q: Which layer handles ports?
A: Transport layer.
Q: Which layer handles encryption and compression?
A: Presentation layer.
Q: Which layer manages sessions?
A: Session layer.
Q: What are the four TCP/IP layers?
A: Application, Transport, Internet/Network and Network Interface.
Q: What does ARP do?
A: Resolves an IP address to a physical/MAC address.
Q: What does ICMP do?
A: Communicates network/IP delivery problems and control information.
Q: What is TCP?
A: A reliable, connection-oriented transport protocol.
Q: What is UDP?
A: A connectionless, unreliable transport protocol.
---
LECTURE 3 — IP ADDRESSING AND SUBNETTING
1. INTERNET PROTOCOL
IP = Internet Protocol
IP is the protocol/method used to send data from one computer to another over interconnected networks.
Each host has an IP address.
When information such as an email or web page is transmitted, the information is divided into smaller units called packets.
Each packet contains:
- Source address
- Destination address
- Data
---
2. LAYERED DATA TRANSMISSION
During transmission, each layer can add header information to the data.
At the sender:
Data → Header added → More headers added → Transmission
At the receiver:
Transmission → Header processed/removed → Data reconstructed
This allows each layer to perform its own function.
---
3. IPv4 DATAGRAM
An IPv4 datagram consists of:
Header + Payload/Data
The header contains control and addressing information.
Important IPv4 header fields include:
1. Version
2. IHL
3. Type of Service
4. Total Length
5. Identification
6. Flags
7. Fragment Offset
8. TTL
9. Protocol
10. Header Checksum
11. Source Address
12. Destination Address
13. Options
14. Padding
---
4. VERSION
The Version field identifies the IP version being used.
Examples:
- IPv4
- IPv6
---
5. IHL
IHL = Internet Header Length
The lecture describes IHL as the actual length of the header expressed in multiples of 32 bits.
---
6. TYPE OF SERVICE
The Type of Service field is associated with prioritization.
It can be used to indicate how traffic should be treated.
---
7. TOTAL LENGTH
The Total Length field represents the length of:
Header + Payload
---
8. IDENTIFICATION
The Identification field helps the receiver reconstruct the original datagram when fragmentation occurs.
Fragments belonging to the same original datagram can be associated using the identification information.
---
9. FLAGS
The Flags field contains information related to fragmentation.
It indicates whether fragmentation is allowed and whether additional fragments are associated with the current datagram.
---
10. FRAGMENT OFFSET
The Fragment Offset identifies the position of the first byte of a fragment relative to the original datagram.
Fragment offsets are based on multiples of 8 bytes.
All fragments except the last one contain a data size that follows the required multiple-of-8 relationship.
---
11. TTL
TTL = Time To Live
TTL indicates the maximum time a packet can remain in the Internet/network.
Its purpose is to prevent packets from remaining in the network indefinitely.
---
12. PROTOCOL FIELD
The Protocol field identifies the next higher-level protocol carried by the IP datagram.
Examples:
- TCP
- UDP
Therefore:
IP → Protocol field → TCP/UDP
---
13. HEADER CHECKSUM
The Header Checksum is used to help detect corruption in the IPv4 header.
It safeguards the header against transmission errors.
---
14. SOURCE ADDRESS
Contains the Internet-wide IP address of the source host.
---
15. DESTINATION ADDRESS
Contains the Internet-wide IP address of the destination host.
---
16. OPTIONS
The Options field can contain additional information.
The lecture mentions examples including:
- Security
- Source routing
- Loose source routing
---
17. PADDING
Padding consists of extra bits added so that the header reaches the required multiple of 32 bits.
---
18. IPV4 FRAGMENTATION
Fragmentation occurs when an IP datagram is too large to be transmitted through a network link with a smaller maximum transmission size.
The original datagram can be divided into smaller fragments.
Each fragment contains relevant header information.
Important fields for fragmentation/reassembly include:
- Identification
- Flags
- Fragment Offset
---
19. FRAGMENTATION AND REASSEMBLY
Sender/intermediate network
A large datagram can be divided into smaller fragments.
Receiver
The fragments are used to reconstruct the original datagram.
The Identification field allows fragments belonging to the same original datagram to be associated.
The Fragment Offset indicates the location of each fragment's data.
---
20. IP ADDRESS ALLOCATION METHODS
The lecture covers four approaches:
1. Class-based addressing
2. Subnetting
3. Classless addressing
4. Network Address Translation
---
21. CLASS-BASED ADDRESSING
Traditional class-based IPv4 addressing divides addresses into classes.
The major classes relevant to the lecture are:
- Class A
- Class B
- Class C
The address is divided into:
NetID + HostID
The NetID identifies the network.
The HostID identifies a host within that network.
---
22. SPECIAL ADDRESSING RULES
Host ID = all 0s
Identifies the network itself.
This is called the network address.
Net ID = all 0s
Refers to the current/local network.
All bits = 1
Represents a limited broadcast.
It sends a message to everyone on the current network.
Host ID = all 1s
Represents a direct broadcast to all hosts on the specified network.
127.x.x.x
The lecture identifies the Class A 127.x.x.x range as a loopback address used for internal testing on the local machine.
---
23. CLASS A
Class A addresses have a large host portion.
They are designed for very large networks.
For a traditional Class A network:
- The first portion identifies the network.
- The remaining portion identifies hosts.
The lecture requires being able to determine:
- Range of NetIDs
- Number of possible networks
- Number of hosts
- Host address range
---
24. CLASS B
Class B provides a balance between network count and host capacity.
The lecture requires determining:
- Range of NetIDs
- Number of possible networks
- Number of hosts
- Host address range
---
25. CLASS C
Class C provides many networks, with fewer hosts per network.
The lecture requires determining:
- Range of NetIDs
- Number of possible networks
- Number of hosts
- Host address range
---
26. SUBNETTING
Subnetting divides a larger network into smaller logical networks called subnets.
Normally:
IP address = NetID + HostID
With subnetting:
IP address = NetID + SubnetID + Local HostID
---
27. WHY SUBNETTING IS USED
A site may contain multiple LANs.
Instead of assigning a completely different network ID to every LAN, the site can use one overall network address and divide it into subnets.
Some bits from the HostID are borrowed and used as a SubnetID.
Structure
Original:
NetID + HostID
After subnetting:
NetID + SubnetID + Local HostID
---
28. SUBNET MASK
A subnet mask identifies which bits belong to:
- Network/Subnet
- Host
Fundamental rule
1 = Network/Subnet
0 = Host
This is one of the most important rules in subnetting.
---
29. SUBNETTING EXAMPLE
Given:
Network: 128.10.12.0
Mask: 255.255.255.0
Binary mask:
11111111.11111111.11111111.00000000
Therefore:
- First 24 bits → network/subnet portion
- Last 8 bits → host portion
The lecture notes that because this relates to a Class B network, the first two bytes represent the Internet-wide NetID and the next byte can represent the SubnetID.
---
30. SUBNETTING CALCULATION
Example from the lecture:
A campus is assigned:
150.10.0.0
Requirements:
- 50 LANs/subnets
- Up to 1000 hosts per subnet
The task is to determine:
1. IP class
2. Suitable address mask
Step 1 — Determine class
150 falls within the traditional Class B range.
Therefore:
Class B
Step 2 — Determine subnet bits
Need at least 50 subnets.
Find a power of 2 that can provide at least 50:
[
2^5=32
]
Not enough.
[
2^6=64
]
Therefore at least 6 subnet bits are required.
Step 3 — Determine host bits
A subnet must support up to 1000 hosts.
Find the number of host bits required:
[
2^{10}=1024
]
Therefore 10 host bits are required for 1000 hosts.
Step 4 — Determine prefix length
IPv4 has 32 bits.
[
32-10=22
]
Therefore:
/22
The corresponding subnet mask is:
255.255.252.0
---
31. CLASSLESS ADDRESSING
Classless addressing allows the network portion of an IP address to contain any number of bits.
It is represented as:
w.x.y.z/n
where:
n = number of network bits
This is also known as CIDR-style notation.
---
32. CIDR
CIDR allows IP address blocks to be allocated more efficiently.
For example:
/22
means:
- 22 network bits
- 10 host bits
because:
[
32-22=10
]
---
33. CIDR EXAMPLE
An organization requires approximately 1000 host addresses.
The next suitable power of two is:
[
2^{10}=1024
]
Therefore, a block of 1024 addresses can be allocated.
Since 10 bits are needed for hosts:
[
32-10=22
]
Therefore the allocation can be represented as:
/22
Important rule
[
\boxed{\text{Host bits}=32-\text{prefix length}}
]
---
34. CLASSFUL VS CLASSLESS ADDRESSING
Class-Based| Classless
Uses predefined classes| Network portion can be any length
Less flexible| More flexible
Can waste addresses| More efficient address allocation
Uses Class A/B/C boundaries| Uses prefix length such as /22
---
35. NETWORK ADDRESS TRANSLATION
NAT = Network Address Translation
NAT allows a network to use private/local IP addresses internally while using an Internet-facing/global address for communication outside the network.
A NAT gateway/router performs the translation.
---
36. PRIVATE IP ADDRESS BLOCKS IN THE LECTURE
The lecture lists three private address blocks:
- 10.0.0.0 – 10.255.255.255 /8
- 172.10.0.0 – 172.31.255.255 /12
- 192.168.0.0 – 192.168.255.255 /16
The second range appears as 172.10.0.0 in the lecture material.
---
37. NAT TABLE
A NAT router maintains a NAT table.
The table associates:
Local IP + Local Port
with
Global IP + Global Port
Example structure:
Local IP| Local Port| Global IP| Global Port
10.0.0.3| 2750| 168.24.16.7| 4001
The mapping allows returning Internet traffic to be associated with the correct internal host.
---
38. NAT COMMUNICATION PROCESS
Example:
Internal host:
10.0.0.3 : 2750
wants to communicate with a web server:
128.40.19.12 : 80
The NAT router changes the source information to a global address/port.
For example:
10.0.0.3 : 2750
becomes:
168.24.16.7 : 4001
The NAT table records the mapping.
When the web server replies to:
168.24.16.7 : 4001
the NAT router uses its table to determine that the traffic belongs to:
10.0.0.3 : 2750
---
39. IMPORTANT LECTURE 3 FORMULAS
Host bits
[
\boxed{H=32-n}
]
where n is the network prefix.
Number of possible combinations
[
2^{\text{number of bits}}
]
Subnet requirement
If S subnet bits are used:
[
2^S
]
possible subnet combinations are available, subject to the addressing rules being applied.
Host capacity
If H host bits are available:
[
2^H
]
possible bit combinations exist.
---
40. BINARY AND DECIMAL CONVERSION
Binary-decimal conversion is a stated requirement for Lecture 3.
An IPv4 address contains four 8-bit octets.
Each octet uses the binary place values:
128 64 32 16 8 4 2 1
Example:
255
=
128 + 64 + 32 + 16 + 8 + 4 + 2 + 1
Therefore:
11111111 = 255
Similarly:
0 = 00000000
---
LECTURE 3 — KEY EXAM POINTS
- IP definition
- IPv4 datagram
- IPv4 header fields
- Fragmentation
- Reassembly
- Identification
- Flags
- Fragment Offset
- TTL
- Protocol field
- Source/destination addresses
- Class A/B/C
- Network ID
- Host ID
- Network address
- Broadcast addresses
- Subnetting
- Subnet mask
- Binary conversion
- CIDR
- Prefix length
- NAT
- NAT table
- Subnet calculations
- Host calculations
---
LECTURE 3 — FLASHCARDS
Q: What is IP?
A: The Internet Protocol used to send data between hosts across networks.
Q: What is an IPv4 datagram composed of?
A: Header and payload/data.
Q: What does the Version field identify?
A: The IP version.
Q: What does IHL represent?
A: The IPv4 header length in multiples of 32 bits.
Q: What does Total Length contain?
A: The length of the header plus payload.
Q: What does Identification help with?
A: Reassembling fragmented datagrams.
Q: What does Fragment Offset indicate?
A: The position of a fragment relative to the original datagram.
Q: What is the fragment offset based on?
A: Multiples of 8 bytes.
Q: What does TTL mean?
A: Time To Live.
Q: What does the Protocol field identify?
A: The next higher-level protocol, such as TCP or UDP.
Q: What does the Header Checksum protect?
A: The IPv4 header against corruption.
Q: What is subnetting?
A: Dividing a larger network into smaller logical subnets by borrowing host bits.
Q: What is the subnetted IP structure?
A: NetID + SubnetID + Local HostID.
Q: What does 1 mean in a subnet mask?
A: Network/Subnet.
Q: What does 0 mean in a subnet mask?
A: Host.
Q: What does /22 mean?
A: 22 network bits and 10 host bits.
Q: How many host bits are present in /22?
A: 10.
Q: How many addresses are represented by 10 bits?
A: 1024.
Q: What is CIDR?
A: Classless addressing using a variable-length network prefix.
Q: What is NAT?
A: Network Address Translation.
Q: What does a NAT table map?
A: Local IP/port combinations to global IP/port combinations.
---
LECTURE 4 — THE INTERNET, PROTOCOLS, NETWORK EDGE AND NETWORK CORE
1. THE INTERNET
The Internet can be described as a network of networks.
It consists of many interconnected networks, including:
- Home networks
- Mobile networks
- Enterprise networks
- Local/regional ISPs
- National/global ISPs
- Datacenter networks
- Content provider networks
---
2. INTERNET — NUTS-AND-BOLTS VIEW
The physical/network infrastructure contains:
Hosts
Computing devices at the Internet edge.
Examples:
- Computers
- Smartphones
- Internet-connected appliances
- Servers
Packet switches
Devices that forward packets.
Examples:
- Routers
- Switches
Communication links
Examples:
- Fibre
- Copper
- Radio
- Satellite
Bandwidth
The transmission rate of a communication link is commonly described as its bandwidth.
---
3. INTERNET — SERVICE VIEW
From a service perspective, the Internet provides infrastructure and communication services to applications.
Examples include:
- Web
- Streaming video
- Multimedia conferencing
- Games
- E-commerce
- Social media
- Connected appliances
The Internet provides an interface that allows distributed applications to send and receive data using network services.
---
4. INTERNET STANDARDS
Internet standards help ensure that different systems can communicate.
The lecture identifies:
RFC = Request for Comments
IETF = Internet Engineering Task Force
IETF develops and publishes Internet-related standards and specifications through documents such as RFCs.
---
5. PROTOCOLS
A protocol is a set of rules governing communication.
Network protocols define:
1. Format of messages
2. Order of messages
3. Actions taken when messages are transmitted
4. Actions taken when messages are received
Simple definition
Protocol = rules governing communication between network entities
---
6. HUMAN PROTOCOL VS NETWORK PROTOCOL
Human communication also follows protocols.
For example, a person may:
1. Say hello
2. Ask a question
3. Receive a response
Similarly, computers follow defined communication rules.
The difference is that computer network protocols specify communication precisely enough for devices to process messages automatically.
---
7. NETWORK EDGE
The network edge consists primarily of end systems/hosts.
Two major types are:
- Clients
- Servers
Clients
Request services.
Servers
Provide services.
Servers are often located in data centres.
---
8. ACCESS NETWORK
An access network connects an end system to the wider network.
The lecture identifies:
1. Residential access
2. Institutional access
3. Mobile access
Two important characteristics are:
- Transmission rate
- Shared or dedicated access
---
9. RESIDENTIAL ACCESS
Residential users connect their home networks to an Internet service provider.
A home network can include:
- Modem
- Router
- Firewall/NAT
- Ethernet
- Wi-Fi access point
- Wireless devices
These functions are often combined into a single device.
---
10. INSTITUTIONAL/ENTERPRISE NETWORKS
Organizations such as companies and universities use networks to connect many devices.
An enterprise network may contain:
- Switches
- Routers
- Ethernet connections
- Wi-Fi
- Servers
- Wireless devices
---
11. WIRELESS ACCESS
Wireless access allows devices to communicate without a physical cable.
WLAN
Wireless Local Area Network.
Wi-Fi is commonly used for WLAN access.
The lecture associates Wi-Fi with IEEE 802.11.
Cellular networks
Provide wide-area wireless access through mobile operators.
Examples include:
- 4G
- 5G
---
12. PHYSICAL MEDIA
Communication media are divided into:
1. Guided media
2. Unguided media
---
13. GUIDED MEDIA
In guided media, signals travel through a physical medium.
Examples:
- Twisted pair
- Coaxial cable
- Fibre optic cable
---
14. UNGUIDED MEDIA
In unguided media, signals travel through space using electromagnetic/radio waves.
Example:
- Radio
---
15. TWISTED-PAIR CABLE
Twisted-pair cable consists of pairs of copper wires twisted together.
The lecture discusses:
- Category 5
- Category 6
Cat5
Can support Ethernet speeds including 100 Mbps and 1 Gbps according to the lecture.
Cat6
Can support 10 Gbps Ethernet according to the lecture.
---
16. COAXIAL CABLE
Coaxial cable contains two concentric copper conductors.
Characteristics include:
- Copper-based
- Bidirectional communication
- Broadband capability
- Multiple frequency channels can be supported
---
17. FIBRE OPTIC
Fibre optic cable uses light pulses transmitted through glass fibre.
Advantages
- Very high transmission speeds
- 10s to 100s of Gbps according to the lecture
- Low error rate
- Long distances between repeaters
- Resistant to electromagnetic interference
---
18. WIRELESS RADIO
Wireless communication uses electromagnetic radiation/radio waves.
Examples
- Wi-Fi
- Cellular networks
- Satellite communication
Problems affecting radio communication
- Reflection
- Obstruction
- Interference
---
19. NETWORK CORE
The network core consists of a mesh of interconnected routers.
Its main role is to move packets between different parts of the Internet.
---
20. PACKET SWITCHING
In packet switching:
1. An application creates a message.
2. The message is divided into packets.
3. Packets are sent through the network.
4. Routers forward packets from one router to another.
5. Packets eventually reach the destination.
Each packet uses the transmission capacity of the link while it is being transmitted.
---
21. STORE-AND-FORWARD
With store-and-forward packet switching, a router must receive the entire packet before transmitting it onto the next link.
Suppose:
- Packet size = L bits
- Link rate = R bits/second
The time required to transmit the packet onto one link is:
[
\boxed{d_{trans}=\frac{L}{R}}
]
---
22. TRANSMISSION DELAY
Transmission delay is the amount of time required to push all L bits of a packet onto a link operating at R bits/second.
[
\boxed{d_{trans}=\frac{L}{R}}
]
Where:
- (L) = packet length in bits
- (R) = transmission rate in bits/second
---
23. TRANSMISSION DELAY EXAMPLE
Given:
[
L=10\text{ Kbits}
]
[
R=100\text{ Mbps}
]
Convert units:
[
10\text{ Kbits}=10,000\text{ bits}
]
[
100\text{ Mbps}=100,000,000\text{ bits/s}
]
Therefore:
[
d_{trans}=\frac{10,000}{100,000,000}
]
[
d_{trans}=0.0001\text{ s}
]
[
d_{trans}=0.1\text{ ms}
]
Therefore:
Transmission delay = 0.1 ms
---
24. TWO-LINK STORE-AND-FORWARD
If a packet must cross two identical links and propagation delay is ignored:
First link:
[
\frac{L}{R}
]
Second link:
[
\frac{L}{R}
]
Total:
[
\boxed{\frac{2L}{R}}
]
For three identical links:
[
\boxed{\frac{3L}{R}}
]
In general, for (N) identical links:
[
\boxed{\frac{NL}{R}}
]
when only transmission delay is considered.
---
25. QUEUEING DELAY
Packets may have to wait before they can be transmitted.
This occurs when packets arrive at a link faster than the link can transmit them.
Process
Arrival rate > transmission rate
↓
Packets wait
↓
Queue grows
↓
Router buffer may become full
↓
Packets may be dropped
---
26. PACKET LOSS
If the router's buffer becomes full, incoming packets may be discarded.
Therefore:
High traffic → queueing → buffer fills → packet loss
---
27. FORWARDING
Forwarding is a local action.
It moves an arriving packet from:
Input link → appropriate output link
The router examines information such as the destination address and uses a forwarding table to determine the appropriate output.
Simple definition
Forwarding = moving the packet to the next appropriate output link
---
28. ROUTING
Routing is a global action.
It determines the path packets should take from the source to the destination.
Routing uses routing algorithms and routing information.
Simple definition
Routing = determining the overall path
---
29. FORWARDING VS ROUTING
Forwarding| Routing
Local action| Global action
Moves packet to output link| Determines source-destination path
Happens at routers| Uses routing algorithms
Uses forwarding table| Produces/updates routing information
Memory aid
Routing = path
Forwarding = next step
---
30. CIRCUIT SWITCHING
Circuit switching is an alternative to packet switching.
In circuit switching, resources are allocated and reserved for a communication session/call.
The lecture describes:
- Dedicated resources
- No sharing of reserved resources during the call
- Circuit-like guaranteed performance
Traditional telephone networks commonly used circuit switching.
---
31. CIRCUIT SWITCHING
Suppose a communication path consists of several links.
A circuit is established between the source and destination.
Resources on each link are reserved for that communication.
Even when the user is not actively transmitting, the reserved circuit resources remain allocated.
---
32. FDM
FDM = Frequency Division Multiplexing
In FDM, the available frequency range is divided into separate frequency bands.
Each communication session is allocated its own frequency band.
Memory aid
FDM → Frequency
---
33. TDM
TDM = Time Division Multiplexing
In TDM, time is divided into slots.
Each communication session receives one or more periodic time slots.
Memory aid
TDM → Time
---
34. FDM VS TDM
FDM| TDM
Divides frequency| Divides time
Each call gets frequency band| Each call gets time slot(s)
Can transmit within assigned band| Transmits during assigned time slots
---
35. PACKET SWITCHING VS CIRCUIT SWITCHING
Packet switching
- Resources are shared
- Good for bursty data
- No permanent dedicated circuit
- Packets can queue
- Packet loss can occur when buffers fill
- Efficient resource sharing
Circuit switching
- Resources are reserved
- Dedicated resources
- Circuit-like guaranteed performance
- Resources can remain idle during periods of no activity
- Traditionally used in telephone networks
---
36. BURSTY DATA
Packet switching is especially useful for bursty data.
Bursty data means that an application may:
- Send a lot of data at one time
- Send little or no data at another time
Packet switching allows network resources to be shared between users instead of permanently reserving resources for each user.
---
37. INTERNET STRUCTURE
The Internet consists of many interconnected networks.
Examples include:
- Home networks
- Mobile networks
- Enterprise networks
- Local/regional ISPs
- National/global ISPs
- Datacenter networks
- Content provider networks
---
38. ISP
ISP = Internet Service Provider
An ISP provides Internet connectivity to customers.
Different levels of networks can connect to each other to create the global Internet.
---
39. IXP
IXP = Internet Exchange Point
An Internet Exchange Point provides infrastructure through which different networks can interconnect and exchange traffic.
---
40. PEERING
Peering occurs when networks establish direct connections to exchange traffic.
This allows networks to exchange traffic without necessarily sending all traffic through another intermediary network.
---
41. TIER-1 ISP
The lecture discusses Tier-1 commercial ISPs as highly connected networks with national or international coverage.
They form an important part of the Internet's interconnected structure.
---
42. CONTENT PROVIDER NETWORKS
Large content providers can operate their own networks to connect their data centres.
This allows them to efficiently distribute services and content to users.
---
43. BITS AND BYTES
A bit is a binary digit.
A byte contains 8 bits.
[
\boxed{1\text{ Byte}=8\text{ bits}}
]
The lecture gives:
[
1\text{ Kb}=10^3\text{ bits}
]
[
1\text{ Mb}=10^6\text{ bits}
]
[
1\text{ Gb}=10^9\text{ bits}
]
For bytes:
[
1\text{ KB}=2^{10}\text{ bytes}
]
[
1\text{ MB}=2^{20}\text{ bytes}
]
[
1\text{ GB}=2^{30}\text{ bytes}
]
[
1\text{ TB}=2^{40}\text{ bytes}
]
Important exam distinction
b = bit
B = Byte
---
LECTURE 4 — KEY EXAM POINTS
- Definition of Internet
- Network of networks
- Nuts-and-bolts view
- Service view
- Protocol
- Internet standards
- RFC
- IETF
- Network edge
- Network core
- Access networks
- Residential networks
- Institutional networks
- Mobile networks
- Guided media
- Unguided media
- Twisted pair
- Coaxial
- Fibre optic
- Wireless radio
- Packet switching
- Store-and-forward
- Transmission delay
- Queueing delay
- Packet loss
- Forwarding
- Routing
- Circuit switching
- FDM
- TDM
- Packet switching vs circuit switching
- ISP
- IXP
- Peering
- Bits vs bytes
---
LECTURE 4 — FLASHCARDS
Q: What is the Internet?
A: A network of interconnected networks.
Q: What is a packet switch?
A: A device such as a router or switch that forwards packets.
Q: What is bandwidth?
A: The transmission rate of a communication link.
Q: What is the network edge?
A: The part of the Internet containing end systems/hosts such as clients and servers.
Q: What is the network core?
A: A mesh of interconnected routers.
Q: What is an access network?
A: A network connecting an end system to the wider network.
Q: What are guided media?
A: Transmission media in which signals travel through a physical medium.
Q: Give examples of guided media.
A: Twisted pair, coaxial cable and fibre optic cable.
Q: What is unguided media?
A: Transmission media where signals travel through space, such as radio.
Q: What does fibre optic use to transmit data?
A: Light pulses.
Q: What are advantages of fibre optic?
A: Very high speed, low error rate and resistance to electromagnetic interference.
Q: What is packet switching?
A: Dividing messages into packets and forwarding them through the network.
Q: What is store-and-forward?
A: A router receives the entire packet before forwarding it to the next link.
Q: What is transmission delay?
A: The time required to transmit all packet bits onto a link.
Q: What is the transmission delay formula?
A: (d_{trans}=L/R).
Q: What does L represent?
A: Packet length in bits.
Q: What does R represent?
A: Transmission rate in bits per second.
Q: What happens when arrival rate exceeds transmission rate?
A: Packets queue and may eventually be lost if the buffer fills.
Q: What is forwarding?
A: A local action that moves a packet from an input link to an appropriate output link.
Q: What is routing?
A: Determining the overall path packets take from source to destination.
Q: What is circuit switching?
A: Reserving dedicated resources for a communication session.
Q: What does FDM stand for?
A: Frequency Division Multiplexing.
Q: What does TDM stand for?
A: Time Division Multiplexing.
Q: What does FDM divide?
A: Frequency.
Q: What does TDM divide?
A: Time.
Q: What is an ISP?
A: Internet Service Provider.
Q: What is an IXP?
A: Internet Exchange Point.
Q: What is peering?
A: Direct interconnection between networks for exchanging traffic.
Q: How many bits are in one byte?
A: 8 bits.
Q: What is the difference between b and B?
A: b represents bits; B represents bytes.
---
MASTER REVISION TABLE
Concept| Essential meaning
Computer network| Connected devices sharing resources/communicating
NIC| Hardware interface for network connection
MAC| Local hardware/network-interface address
IP| Logical network address
Port| Identifies application/service
Hub| Broadcasts to all ports
Switch| Forwards toward intended destination
Router| Connects different networks
Simplex| One-way
Half-duplex| Two-way, one direction at a time
Full-duplex| Two-way simultaneously
Client/server| Central server provides services
P2P| Peers share resources without dedicated server
PAN| Personal area
LAN| Local area
MAN| Metropolitan area
WAN| Wide area
CIA| Confidentiality, Integrity, Availability
OSI| Seven-layer reference model
Physical| Bits/signals
Data Link| Frames/physical addressing/error/access control
Network| IP/routing/packets
Transport| Ports/segments/end-to-end communication
Session| Session management
Presentation| Translation/encryption/compression
Application| Network applications/services
TCP| Reliable, connection-oriented
UDP| Connectionless, unreliable
ARP| IP → physical/MAC address
ICMP| Network/IP control and error information
IPv4 datagram| Header + payload
Subnetting| Dividing network into subnets
Subnet mask| 1 = network/subnet, 0 = host
CIDR| Variable-length network prefix
NAT| Translates private/local and global addresses
Internet| Network of networks
Network edge| Hosts/end systems
Network core| Interconnected routers
Packet switching| Shared packet-based transmission
Store-and-forward| Entire packet received before forwarding
Transmission delay| (L/R)
Queueing| Packets wait for transmission
Forwarding| Local next-hop packet movement
Routing| Overall path determination
Circuit switching| Reserved dedicated resources
FDM| Frequency division
TDM| Time division
ISP| Internet Service Provider
IXP| Internet Exchange Point
Peering| Direct traffic exchange between networks
---
MOST IMPORTANT FORMULAS
Transmission Delay
[
\boxed{d_{trans}=\frac{L}{R}}
]
where:
- (L) = packet size in bits
- (R) = transmission rate in bits/second
---
Multiple Identical Links
For (N) identical links when considering only transmission delay:
[
\boxed{d=\frac{NL}{R}}
]
---
IPv4 Host Bits
[
\boxed{\text{Host bits}=32-\text{network bits}}
]
For CIDR /n:
[
\boxed{\text{Host bits}=32-n}
]
---
Number of Bit Combinations
[
\boxed{2^n}
]
where (n) is the number of bits.
---
IPv4 Address Size
[
\boxed{32\text{ bits}}
]
---
IPv6 Address Size
[
\boxed{128\text{ bits}}
]
---
Byte Conversion
[
\boxed{1\text{ Byte}=8\text{ bits}}
]
---
FINAL MEMORY MAP
LECTURE 1
Devices + communication + network types + topology + security
MAC → IP → Port
Hub → Switch → Router
Simplex → Half-duplex → Full-duplex
Client/Server → P2P
PAN → LAN → MAN → WAN
CIA → Confidentiality, Integrity, Availability
---
LECTURE 2
Layers + models + protocols
OSI:
Physical → Data Link → Network → Transport → Session → Presentation → Application
TCP/IP:
Network Interface → Internet → Transport → Application
Network → IP/routing
Transport → TCP/UDP/ports
Presentation → encryption/compression/translation
Application → HTTP/DNS/DHCP/SMTP
---
LECTURE 3
IP addressing + subnetting + NAT
IPv4 = 32 bits
Datagram = Header + Payload
Fragmentation:
Identification + Flags + Fragment Offset
Subnet:
NetID + SubnetID + HostID
Subnet mask:
1 = Network/Subnet
0 = Host
CIDR:
w.x.y.z/n
Host bits:
32 − n
NAT:
Private IP/Port ↔ Global IP/Port
---
LECTURE 4
Internet + protocols + edge + core + switching
Internet = Network of networks
Edge = Hosts
Core = Routers
Protocol = Rules
Packet switching = Packets + shared resources
Store-and-forward = Receive entire packet → forward
Transmission delay:
[
\frac{L}{R}
]
Forwarding = local action
Routing = global path
Circuit switching = reserved resources
FDM = Frequency
TDM = Time
---
HIGH-PRIORITY EXAM TOPICS
Very High Priority
1. OSI seven layers and functions
2. TCP/IP model and comparison with OSI
3. TCP vs UDP
4. IPv4 header fields
5. IPv4 fragmentation and reassembly
6. Subnetting calculations
7. Subnet masks
8. CIDR calculations
9. NAT and NAT tables
10. Transmission delay calculations
11. Forwarding vs routing
12. Packet switching vs circuit switching
13. Simplex/half-duplex/full-duplex
14. Hub vs switch
15. Client/server vs P2P
High Priority
16. MAC vs IP vs Port
17. PAN/LAN/MAN/WAN
18. Network topologies
19. CIA security principles
20. Network edge/core
21. Guided vs unguided media
22. Fibre/twisted pair/coax
23. FDM vs TDM
24. ARP
25. ICMP
26. ISP/IXP/peering
Formulas to Memorise
[
\boxed{d_{trans}=\frac{L}{R}}
]
[
\boxed{\text{Host bits}=32-n}
]
[
\boxed{2^n=\text{number of bit combinations}}
]
[
\boxed{1\text{ Byte}=8\text{ bits}}
]