COMP1055 Lecture 13: Routing and IP Study Notes

University of Nottingham - Routing and IP - Lecture 13 Notes
Overview

  • Instructor: Dr. Steven Bagley, University of Nottingham

  • Lecture focus: Routing and Internet Protocol (IP)

Recap: Packet Forwarding

  • Introduces the fundamental aspect of networking where data packets are forwarded from one network to another until they reach their destination.

    • Network 1 and Network 2 are highlighted as the two distinct networks involved in packet forwarding.

    • The references to various layers (5 - Application, 4 - Transport, 3 - Internet, 2 - Network Interface, 1 - Physical) within the context of communication between these networks illustrate the layered architecture of networking.

Internet Protocol (IP)

Basic Functions of IP

  • Definition of IP: Internet Protocol facilitates communication across diverse networks. It is primarily considered a layer 3 protocol designed for routing.

  • Functions:

    • Addressing: Identifies both the machine and the respective network on which the machine is located, allowing IP to deliver packets properly.

    • Fragmentation: IP can break down larger packets into smaller units suitable for transmission across different networks.

    • Connectionless Nature: IP sends datagrams without guaranteeing the order of delivery or even ensuring that they arrive.

Packet Structure and Addressing

  • IP Address Format: An IPv4 address is composed of 4 bytes, often represented in a dotted decimal format (e.g., 128.243.28.210). Though it notionally allows for approximately 4 billion devices, IPv4 has reached exhaustion in available addresses.

    • Classes of IP addresses:

    • Class A: Huge networks with up to 16.7 million addresses (range: 1.x.x.x to 127.x.x.x).

    • Class B: Networks accommodating 65,536 addresses (range: 128.0.x.x to 191.255.x.x).

    • Class C: Networks with 256 addresses (range: 192.0.0.x to 223.255.255.x).

    • Class D: Reserved for multicast (range: 224.0.0.0 to 239.255.255.255).

    • Class E: Reserved for experimental use (range: 240.0.0.0 to 255.255.255.255).

  • Limitations of Classful Addressing:

    • Inflexibility: Organizations often received more addresses than needed, leading to significant waste.

    • Scalability: The fixed sizes did not cater well to mid-sized organizations that required more than 256 but fewer than 65,536 addresses.

Classless Inter-Domain Routing (CIDR)

  • Introduced to reduce exhaustion of IP addresses and the rapid increase in routing tables.

  • CIDR employs a variable-length prefix specifying the network part of an address (e.g., 10.0.0.1/8, 192.168.1.1/24) rather than fixed 8-bit segments.

  • Benefits of CIDR:

    • Efficiency: Allows for much finer control over address allocation, minimizing waste.

    • Aggregation: Enables "route summarization," where a single routing table entry can represent a large block of networks, reducing the memory and processing burden on routers.

Transition to IPv6

  • The limitations of IPv4 led to the development of IPv6 in the 1990s, which utilizes a 128-bit address space.

  • Benefits of IPv6:

    • Scalability: Provides approximately 3.4 \times 10^{38} addresses, effectively solving the exhaustion problem.

    • Efficiency: Simplified header format improves processing speed in routers.

    • Security: Built-in support for IPsec (Internet Protocol Security).

  • Limitations of IPv6:

    • Compatibility: Not backward-compatible with IPv4, requiring transition mechanisms like tunneling or dual-stacking.

    • Complexity: Longer addresses are harder for humans to memorize and manage manually.

Routing

Nature of Routing

  • Routing is fundamentally a layer 3 process focused on determining the optimal path for data packets across networks.

  • The Store-and-Forward Mechanism:

    • Benefit: Ensures error checking can happen before a packet is sent onto the next link.

    • Limitation: Adds latency, as the entire packet must be received before it can be processed and forwarded.

Routing Tables and Source Independence

  • Each machine uses a routing table that connects destination addresses with the next hop.

  • Source Independence:

    • Benefit: Simplifies routing logic; routers only need to know where a packet is going, not where it came from.

    • Limitation: Can make certain types of traffic engineering or security filtering (like preventing spoofing) more difficult without additional mechanisms.

Reliability of IP

  • IP is classified as a best-effort protocol.

  • Benefits of Best-Effort Delivery:

    • Low Overhead: No overhead for acknowledgments or retransmissions within the IP layer itself, maximizing speed for real-time applications.

    • Simplicity: Routers are kept simple and fast because they do not track the state of connections.

  • Limitations of Best-Effort Delivery:

    • Unreliability: Packets can be lost, duplicated, or arrive out of order, necessitating higher-layer protocols like TCP for reliable data transfer.