Show understanding of Ethernet and how collisions are detected and avoided

2.1 Networks – The Internet

2.1.1 Purpose and Benefits of Networking

• Resource sharing – e.g., a single printer or a central file server can be used by all computers in a school.
• Data sharing & collaboration – email, file transfer and real‑time editing of documents allow students and staff to work together from different rooms.
• Scalability – new users or devices can be added without rewiring the whole system; a growing college can simply plug in extra switches.
• Cost efficiency – centralised hardware (servers, backup devices) reduces the number of duplicate devices and lowers maintenance expenses.
• Reliability & redundancy – multiple paths or backup links keep the network running even if one cable or device fails.

2.1.2 LAN vs. WAN

Think‑prompt: *Which type of network would you choose for a multinational corporation and why?* (Consider distance, control, cost and performance.)

2.1.3 Network Models

• Client‑Server – Dedicated servers host services (web, file, email) that client devices request.
• Peer‑to‑Peer (P2P) – Every device can act as both client and server; resources are shared directly between peers.
• Thin vs. Thick (Fat) Clients

  • Thin client – minimal processing; applications and data run on a central server.
  • Thick client – runs applications locally and stores data on its own hard‑disk; needs more powerful hardware.

2.1.4 Network Topologies

2.1.5 Wired vs. Wireless LANs

2.1.6 Cloud Computing Concepts

• Infrastructure as a Service (IaaS) – virtualised hardware (e.g., AWS EC2).
• Platform as a Service (PaaS) – development platforms (e.g., Google App Engine).
• Software as a Service (SaaS) – applications delivered over the Internet (e.g., Office 365).
• Data‑centres that host cloud services rely on high‑speed WAN links and use Ethernet extensively within server racks.

2.1.7 LAN Hardware Components

2.1.8 Ethernet Standards, Speeds & Cabling

2.1.9 Bit‑Streaming Concepts

• Serial transmission – bits are sent one after another over a single pair (the method used on Ethernet copper cables).
• Parallel transmission – multiple bits travel simultaneously on separate wires (used inside NICs and on older computer buses).
• Encoding schemes

  • Manchester – used in 10 Mbps Ethernet; combines clock and data.
  • 4B/5B, 8B/10B – used on Fast, Gigabit and 10‑Gigabit Ethernet to provide enough transitions for clock recovery and to keep the DC balance.

2.1.10 The World Wide Web vs. The Internet

• Internet – Global network of interconnected networks that uses the TCP/IP protocol suite.
• World Wide Web (WWW) – An application‑layer service that uses HTTP/HTTPS to exchange hyper‑text documents over the Internet.
• Other common Internet services: email (SMTP), file transfer (FTP), remote login (SSH).

2.1.11 IP Addressing & Subnetting

IPv4 Example

IP address: 192.168.12.0/24
Network part: 192.168.12
Host part: 0‑255 (254 usable hosts)
Broadcast address: 192.168.12.255

IPv6 Example

IP address: 2001:0db8:85a3::8a2e:0370:7334/64
Network prefix: 2001:0db8:85a3::/64
Host identifier: 8a2e:0370:7334

• Public vs. Private IPv4 – 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16 are reserved for private use.
• Static vs. Dynamic allocation – Static: manually configured; Dynamic: assigned by DHCP.
• Subnet mask – Determines the size of the network; e.g., 255.255.255.0 for a /24 network.

2.1.12 DNS and URL Resolution

1. User types http://www.example.com in a browser.
2. Browser checks its local DNS cache; if the address is missing, it contacts a recursive DNS resolver.
3. The resolver follows the hierarchy: root server → TLD server (.com) → authoritative server for example.com.
4. The resolver returns the IP address (e.g., 93.184.216.34) to the browser.
5. The browser opens a TCP connection to that IP on port 80 (or 443 for HTTPS) and sends an HTTP request.

2.1.13 OSI and TCP/IP Model Overview

2.1.14 Ethernet Frame Structure

2.1.15 Collision Detection – CSMA/CD (Half‑Duplex)

1. Carrier Sense – A station listens to the medium; if it is idle, the station may start transmitting.
2. Multiple Access – All stations share the same physical medium.
3. Collision Detection

   • During transmission the station monitors the voltage on the cable.
   • A deviation from the expected signal indicates that another station is transmitting at the same time.
   • On detection the station aborts the current frame.

4. Jam Signal – 32 bits of a predefined pattern are sent to ensure every station recognises the collision.
5. Binary Exponential Back‑off

   Wait time = Slot time × Random(0, 2k – 1)

   where k = number of collisions for that frame (max = 10). Slot time = 51.2 µs for 10 Mbps Ethernet.

6. Retransmission – After the back‑off period the station repeats the carrier‑sense step.

Minimum Frame Size & Propagation Delay

To guarantee that a transmitting station can detect a collision, the frame must keep the medium busy for at least twice the maximum propagation delay (τ).

Formula: Minimum frame size (bits) = 2 × τ × Data rate

Example (10 Mbps Ethernet, max cable 500 m, propagation speed ≈ 2 × 10⁸ m/s):

τ = 500 m / (2 × 10⁸ m/s) = 2.5 µs
Minimum size = 2 × 2.5 µs × 10 Mbps = 50 bits ≈ 64 bytes (512 bits) in practice

2.1.16 Collision Avoidance – Full‑Duplex Ethernet

• Separate transmit and receive pairs (or separate wavelengths on fibre) give each direction a dedicated, point‑to‑point link.
• Because the two directions are physically isolated, a station never sees another station’s signal on its receive pair – collisions are impossible.
• Full‑duplex operation requires a switch (or a router with Ethernet ports); hubs cannot provide full‑duplex links.
• With full‑duplex, CSMA/CD is disabled and the network can operate at the maximum speed of the link (e.g., 1 Gbps on 1000BASE‑TX).