Computer Networks – Unit: 3 Network Layer

1. Point-to-Point Networks

Definition:
A point-to-point network connects exactly two nodes (devices) directly using a dedicated
communication link.

Example: Serial cable connection between two computers.

Keywords: direct link, dedicated connection, two nodes

2. Logical Addressing

Definition:
Logical addressing is used to uniquely identify each device on a network. The most common logical
address is the IP address (Internet Protocol address).

Why Needed?
Physical (MAC) addresses are fixed to hardware, while logical addresses (IP) help in identifying
location in the network.

3. Basic Internetworking Components

a. IP (Internet Protocol)

Definition:
IP is a network layer protocol used to deliver packets from the source to destination across multiple
networks.

Types:

• IPv4: 32-bit address (e.g., 192.168.1.1)

• IPv6: 128-bit address (e.g., 2001:0db8::1)

b. CIDR (Classless Inter-Domain Routing)

Definition:
CIDR is a method for efficient IP address allocation and route summarization. It replaces classful
addressing.

CIDR Notation: IP address + suffix → 192.168.1.0/24

• /24 means 24 bits for network; 8 bits for host.

Benefit: Reduces waste of IP addresses.

c. ARP (Address Resolution Protocol)

Definition:
ARP maps IP addresses to MAC addresses.

Use: When a host knows IP but needs to find the corresponding MAC address to send the data.

d. RARP (Reverse ARP)

Definition:
RARP maps MAC addresses to IP addresses.

Use: Useful for diskless computers that know their MAC but need IP during boot.

e. DHCP (Dynamic Host Configuration Protocol)

Definition:
DHCP dynamically assigns IP addresses to devices on a network.

Process (DORA):

1. Discover

2. Offer

3. Request

4. Acknowledge

Use: Automates the IP configuration process.

f. ICMP (Internet Control Message Protocol)

Definition:
ICMP is used for error reporting and diagnostics in IP networks.

Use: Tools like ping and traceroute use ICMP.

4. Routing, Forwarding, and Delivery

a. Routing

Definition:
The process of selecting the best path for data to travel from source to destination.
Routing Table: Maintains available paths and their costs.

b. Forwarding

Definition:
Refers to the actual process of sending a packet from the router to the next hop based on routing
table.

c. Delivery Types:

• Direct Delivery: When sender and receiver are on the same network.

• Indirect Delivery: Uses routers when they are on different networks.

Diagram – Routing vs Forwarding

css

CopyEdit

[Host A] → [Router 1] → [Router 2] → [Host B]

↑ ↑

Routing Forwarding

5. Static and Dynamic Routing

a. Static Routing

Definition:
Manual configuration of routing tables by network admin.

• Pros: Simple, secure.

• Cons: Not scalable, no automatic updates.

b. Dynamic Routing

Definition:
Routers automatically learn/update routes using routing algorithms and protocols.

• Pros: Scalable, auto-updates.

• Cons: More complex.

 6. Routing Algorithms and Protocols

a. Distance Vector Routing

• Each router shares its routing table with neighbors periodically.

• Uses Bellman-Ford algorithm.

• Example: RIP (Routing Information Protocol)

b. Link State Routing

• Routers exchange link-state information with all routers.

• Each router builds a full map of network.

• Uses Dijkstra’s algorithm.

• Example: OSPF (Open Shortest Path First)

c. Path Vector Protocol

• Used in inter-domain routing.

• Routers advertise entire path information.

• Example: BGP (Border Gateway Protocol)

Diagram – Difference between Distance Vector and Link State

less

CopyEdit

Distance Vector:

[Router A] ← [Router B] → [Router C]

(Sends entire table to neighbors)

Link State:

[Router A] → Broadcasts link info → All Routers

(Builds full network map)

7. Congestion Control Algorithms

Definition:
Congestion occurs when network traffic exceeds capacity → leads to delay, packet loss.

a. Leaky Bucket Algorithm

• Fixed output rate regardless of incoming rate.

• If bucket (buffer) overflows → packets are discarded.

b. Token Bucket Algorithm

• Tokens are generated at fixed rate.

• To send a packet, you need a token → allows bursty traffic.

c. Load Shedding

• Drops packets when the system is overloaded.

8. IPv6

Definition:
IPv6 is the next-generation Internet Protocol designed to replace IPv4.

Features of IPv6:

• 128-bit address (compared to 32-bit in IPv4)

• Supports 2^128 addresses

• No need for NAT

• Simplified header

• Integrated security (IPSec)

• Auto-configuration of IP addresses

IPv4 vs IPv6 Diagram

arduino

CopyEdit

IPv4: 32-bit → 192.168.1.1

IPv6: 128-bit → 2001:0db8:85a3::8a2e:0370:7334

Summary Keywords for 7-Mark Answers:

• Logical Addressing: Unique IP, CIDR for subnetting.

• ARP/RARP: Resolve IP ↔ MAC.

• DHCP: DORA – dynamic IP config.

• ICMP: Ping, traceroute.

• Routing vs Forwarding: Route = path, Forward = actual send.

• Static vs Dynamic Routing: Manual vs Auto.

• Routing Algorithms: DV (RIP), LS (OSPF), PV (BGP).

• Congestion Control: Leaky Bucket, Token Bucket.

• IPv6: 128-bit, more addresses, no NAT.