Introduction to Network Engineering Notes

Introduction to Network Engineering

IP Service Model

  • Internetwork: Refers to an arbitrary collection of networks interconnected to provide a host-to-host packet delivery service.

    • Purpose of Network Layer:

    • Solve the problem of host-to-host delivery.

    • Facilitate routing packets over networks.

    • Definition: An internetwork is often referred to as a "network of networks" as it comprises multiple smaller networks.

Link Layer vs Network Layer Delivery

  • Link Layer Delivery: Involves hop-to-hop delivery, focusing on transferring packets between directly connected devices.

    • Structures include physical and data link layers.

  • Network Layer Delivery:

    • Involves end-to-end delivery, concerning the overall journey of packets from the source host to the destination.

    • Comparison of Layers:

      • Link Layer: Responsible for data transfer between neighbor devices.

      • Network Layer: Responsible for the path across different networks.

IP Service Model Characteristics

  • Unique Addressing Scheme: Provides a method to identify all hosts in the internetwork.

  • Datagram Approach: Used for data delivery.

  • Connectionless Service: Does not establish a dedicated connection.

  • Unreliable Delivery:

    • If a packet is lost, corrupted, or misdelivered, the network layer takes no corrective action.

    • Characteristics include:

    • Best-effort delivery with no guarantees.

    • No flow control.

    • No error control.

Internet Protocol Version 4 (IPv4)

  • Overview: The delivery mechanism employed by TCP/IP protocols.

IPv4 Datagram Format

  • Structure:

    • Header: 20-60 bytes

    • Data: 20-65,536 bytes

  • Header Fields:

    • VER: 4 bits (IP version)

    • HLEN: 4 bits (Header Length)

    • Service: 8 bits (Type of Service)

    • Total Length: 16 bits (Total length of the datagram including header)

    • Identification: 16 bits (Used for fragmentation)

    • Flags: 3 bits (Indicate fragmentation possibilities)

    • Fragmentation Offset: 13 bits (Position of fragment in datagram)

    • Time to Live (TTL): 8 bits (Limits the lifespan of a datagram)

    • Protocol: 8 bits (Upper-layer protocol encapsulated)

    • Header Checksum: 16 bits (Error checking)

    • Source/Destination IP Address: 32 bits each

    • Options: 32 bits (Optional fields for routing, management, etc.)

Characteristics of the IPv4 Header

  • Fixed and Optional Parts: The header consists of a fixed part of 20 bytes plus optional fields.

  • Fields Explanation:

    • Version: Specifies the version of IP.

    • Header Length (HLEN): Indicates the length of the header.

    • Type of Service: Enables differentiated treatment of packets based on application needs.

    • Total Length: Indicates the entire datagram size.

    • Identification: Segments related to a datagram are tagged with the same identification number.

    • Flags:

    • Denote if fragmentation is allowed.

    • Indicate fragment position (first, middle, last).

    • Offset: Shows the relative position of fragments within the datagram.

    • TTL: The maximum number of hops a datagram is allowed to traverse.

    • Protocol: Codes for the upper-layer protocol used (e.g., TCP, UDP).

    • Header Checksum: Helps in the detection of errors.

    • Source/Destination Address: Identifies where the datagram originated and where it is going.

    • Options: Fields to manage routing and protocols.

Maximum Transfer Unit (MTU)

  • Definition: Maximum layer 3 packet size that can be forwarded out of an interface.

  • Basis: Determined by the data link layer protocol.

  • Default MTU: Commonly set to 1500 bytes for Ethernet.

  • Importance of MTU:

    • If a packet exceeds the MTU size, it must be fragmented to fit within the allowable size.

Fragmentation

  • Purpose: Required when the MTU of the interface is smaller than the packet size.

  • Mechanism: Splits the packet into smaller fragments for transmission.

  • Offset: Shows the position of fragments based on the overall datagram size in units of 8 bytes.

IP Addressing

  • Definition: Each TCP/IP node has a unique 32-bit logical IP address.

  • Total Address Space: 2^32 or 4,294,967,296 (with 3,706 million usable).

  • Structure: An IP address consists of a network number and a host number.

  • Dotted-Decimal Notation: Format for representing IP addresses.

Classful Addressing

  • Address Classes:

    • Class A: First byte starts with 0 (0-127).

    • Class B: First byte starts with 10 (128-191).

    • Class C: First byte starts with 110 (192-223).

    • Class D: First byte starts with 1110 (224-239) (Multicast).

    • Class E: First byte starts with 1111 (240-255) (Reserved for future use).

Reserved Addresses

  • Block Ranges:

    • 0.0.0.0 to 0.255.255.255: "This" Network

    • 10.0.0.0 to 10.255.255.255: Private Use Networks

    • 127.0.0.0 to 127.255.255.255: Loopback Address

    • 192.168.x.x: Private Network Used

    • 240.0.0.0 to 255.255.255.255: Reserved for future use.

Network and Broadcast Addresses

  • Network Number: Represents a network in a 32-bit format, with the host portion zeros.

  • Broadcast Address: Used to address all hosts in a network, with the host portion composed entirely of ones.

  • Example: For IP Address 8.1.4.5:

    • Network address: 8.0.0.0

    • Broadcast address: 8.255.255.255.

Subnetting and Masking

  • Subnet Mask: Helps specify the network ID and host ID.

  • Bitwise Operation: Performing a bitwise AND between the IP address and subnet mask results in the network ID.

Nodes with Multiple Addresses

  • Definition: A node's internet address indicates its connection to a network, including both network (netid) and host parts (hostid).

  • Requirement for Routers: Devices such as routers, which connect multiple networks, need to have multiple addresses (one for each connected network).

Private Networks

  • Definition: Organizations that require IP addresses solely for internal communication without requiring direct internet access may use any network number.

  • RFC 1918: Defines networks that are reserved and may be used internally (Private Addressing) without valid registered IP addresses.

  • NAT (Network Address Translation): A method proposed for allowing hosts on private networks to communicate over the internet.

Advantages and Disadvantages of Private Addressing

  • Advantages: Saves IP address space; provides internal identification without requiring external internet access.

  • Disadvantages: NAT must map private addresses to registered external addresses to enable internet communication.

NAT Implementation

  • Example: Shows how NAT translates private addresses to public addresses:

    • Devices with private addresses (e.g., 172.18.3.x) interact via NAT router connected to the internet.

Dynamic Host Configuration Protocol (DHCP)

  • Function: DHCP dynamically assigns IP addresses to hosts.

  • Process:

    • Host broadcasts a “DHCP discover” message.

    • DHCP server replies with a “DHCP offer” message.

    • Host sends a