Internetworking and Networking Concepts
INTRODUCTION
This section discusses the basic principles of networking, specifically the method of how data is transmitted in packets. This transmission method enables multiplexing of data from multiple machines onto one network channel.
Data Transmission: Data is split into packets, allowing it to travel across the network while coexisting with data from other machines.
Benefits: Allows for efficient use of bandwidth through multiplexing, as multiple users can share the same physical channel rather than requiring dedicated lines.
Ethernet Packet: Discussion includes examination of what an Ethernet packet looks like and the procedure of sending them across a local network.
Network Scaling: A brief mention of challenges encountered when networks grow larger, with a focus on packet behavior under these conditions.
NETWORKING AT SCALE
In this section, the dynamics of larger network setups, particularly Local Area Networks (LANs), are explored.
Small Scale Networks: Focuses on issues faced as the number of computers increases in a network.
Limitations: The risk of collisions—which occur when two devices attempt to send packets simultaneously—is amplified as the network grows, leading to data loss and performance degradation.
Bridges and Switches: These devices assist in segmenting networks to manage traffic flow.
Solutions: By using switches, networks are divided into smaller collision domains, which significantly reduces the frequency of collisions.
The Internet as a Model: Envisions the Internet as comparable to the size of a room or a building.
Limitations: Bridges and switches must maintain awareness of all devices on the network, which becomes computationally expensive and unfeasible on the scale of the global Internet.
ARPANET LOGICAL MAP, JUNE 1979
The ARPANET map displays the interconnections of various computers and networks during 1979, illustrating an early form of network architecture and communication technology. Machines referenced include:
PDP-11s, CDC series, DEC systems, among others.
Locations and institutions where these machines were situated such as MIT, Stanford, and others.
INTERNETWORKING
Internetworking is explained as a solution to avoid building a single massive network by creating multiple smaller networks that are interconnected, resulting in a network of networks.
Definition of Internetworking: The concept is introduced, emphasizing that the Internet consists of heterogeneous networks.
Benefits: Provides flexibility by allowing different types of networking technologies (Ethernet, Wi-Fi, Fiber) to communicate seamlessly as a unified global system.
OSI NETWORK LAYER MODEL
This depth section focuses on the OSI model, featuring seven layers essential for computer networking.
Layers:
Physical Layer: Concerns the physical connection between devices.
Data Link Layer: Manages node-to-node data transfer.
Network Layer: Responsible for packet forwarding—including routing through different networks.
Transport Layer: Ensures complete data transfer.
Session Layer: Manages sessions between applications.
Presentation Layer: Ensures data is in a readable format for the application layer.
Application Layer: Facilitates user interfaces and application services.
Benefits of a Layered Model: Decouples different networking functions, allowing developers to improve technologies at one layer (e.g., upgrading a physical cable) without needing to rewrite software at another layer (e.g., the web browser).
NETWORK INTERFACES
This section discusses the assumption that a computer has only one Network Interface Card (NIC).
Single vs. Multiple NICs: Emphasizes that computers can have multiple NICs such as Ethernet and Wi-Fi, broadening networking capabilities.
Benefits: Redundancy and increased throughput; if one interface fails, the computer can remain connected via another.
LOCAL AREA NETWORKS (LANs)
Describes configurations in LANs, particularly concerning shared mediums and connections.
Shared Medium Concept: In a traditional LAN setup, each computer connects to a shared medium, which facilitates communication with others on that network.
Multiple Interfaces: An illustration shows how a computer with multiple connections can talk to devices on both connected networks.
PACKET FORWARDING
Explores the methods by which data packets can be forwarded between networks.
Rationale for Packet Forwarding: Introduces the concept of routing packets between machines in separate networks.
Router Definition: A router or packet switch is identified as a device specifically set to forward packets.
Benefits: Routers efficiently determine the best path for data across complex topologies, ensuring packets reach remote destinations.
THE INTERNET
Explains the Internet as a network of networks and the mechanics behind packet routing.
Addressing: Addresses are vital for identifying machines and ensuring packets reach their intended destinations.
Machine and Network Addressing: Packets need to specify both the machine and the respective network to properly route.
INTERNET PROTOCOL (IP)
Describes the role of the Internet Protocol in packet management across various networks.
Functionality: Noted as a Layer 3 protocol responsible for addressing, routing, and fragmentation.
IPv4 and IPv6: Discusses the differences between IPv4 and IPv6.
Limitations of IPv4: The 32-bit address space provides only 2^{32} \approx 4,294,967,296 addresses, which have been exhausted by the growth of global devices.
Solutions (IPv6): Expands address length to 128 bits (2^{128}), providing a virtually inexhaustible supply of addresses.
IP ADDRESSES
Details the structuring of IP addresses in networking.
IPv4 Structure: Defined as 4 bytes usually in dotted decimal format.
Address Class System: Includes a breakdown of Class A, B, C, D, and E.
Limitations: Fixed-size classes often led to major inefficiencies, where organizations were allocated far more addresses than they actually needed, accelerating exhaustion.
CLASSLESS INTER-DOMAIN ROUTING (CIDR)
Explains CIDR as a method to combat the rapid exhaustion of IP address space.
Variable Addressing: Introduces CIDR's approach to allow variability in the number of bits defining a network.
Solutions: CIDR replaced the rigid class system, allowing addresses to be assigned in much more precise quantities, which dramatically slowed the rate of IP address depletion.
ADDRESS CONFIGURATION
Touching on how IP addresses can be configured within networks.
Static vs. Dynamic: Defines static IP (manual) and dynamic IP (automatic).
Dynamic Host Configuration Protocol (DHCP):
Benefits: Automates the assignment process, preventing duplicate address errors and reducing the administrative burden on network managers.
Stateless Address Autoconfiguration (SLAAC):
Benefits: Allows IPv6 devices to generate their own address instantly upon connection without needing a central server, enabling true 'plug-and-play' connectivity.