Understanding Internet Architecture and Network Protocols

Internet Fundamentals: A Network of Networks

The Internet is often described as a "network of networks," comprising interconnected Internet Service Providers (ISPs).

Key Components of the Internet

• Hosts: End Systems for Applications

  Hosts are the end systems (clients & servers) that run applications. They connect at the edge of the Internet.

• Packet Switches: Forwarding Data

  Packet switches forward data in chunks called packets. These include:

  • Routers: Forward data between different networks.
  • Switches: Operate within the core of the Internet, not at the edge.

• Communication Links: The Data Pathways

  Communication links connect routers and can be made of:

  • Fiber optic cables
  • Copper wires
  • Radio waves
  • Satellite connections

  The transmission rate of a link is known as its bandwidth.

• Networks: Managed Collections of Devices

  A network is a collection of devices, routers, and links, typically managed by an organization.

• Protocols: Rules for Communication

  Protocols are sets of rules that specify actions, controlling the sending and receiving of messages. Examples include:

  • HTTP (Hypertext Transfer Protocol)
  • Streaming video protocols
  • TCP/IP (Transmission Control Protocol/Internet Protocol)
  • WiFi
  • 4G and 5G cellular technologies

• Network Protocols: Defining Message Exchange

  Network protocols define the format, order of messages sent and received among network entities, and the actions taken upon message transmission.

• Internet Standards: RFCs and IETF

  Internet standards are crucial for interoperability:

  • RFC (Request for Comments): Documents that describe and contain specifications for Internet standards.
  • IETF (Internet Engineering Task Force): The organization responsible for developing and promoting Internet standards.

• Access Networks

  Access networks connect end systems to the edge routers of the Internet.

Delay, Loss, and Throughput in Packet-Switched Networks

Understanding network performance involves analyzing delay, loss, and throughput.

Packet-Switching vs. Circuit-Switching

• Packet-Switching Networks: The Internet Model

  The Internet operates as a packet-switching network, where data is broken into packets and sent independently.

• Circuit-Switching Networks: Reserved Resources

  Examples include traditional telephone networks and ATM (Asynchronous Transfer Mode). In these networks, end-to-end resources are allocated and reserved for the duration of a "call" between source and destination.

Packet Delay: Four Sources

Nodal delay is the total delay a single packet experiences at one network node. It is the sum of four components:

d_nodal = d_proc + d_queue + d_trans + d_prop

• Processing Delay (d_proc)

  The time required to examine the packet's header, check for bit errors, and determine the output link.

• Queueing Delay (d_queue)

  The time a packet waits in a queue for its turn to be transmitted onto the link. This delay depends on the congestion level of the router and the rate at which packets arrive.

• Transmission Delay (d_trans)

  The time it takes for all bits of the packet to be pushed onto the link. This delay depends on the packet's length (L) and the link's transmission rate (R), also known as capacity, bandwidth, or speed (measured in bits per second, bps).

  d_trans = L / R

• Propagation Delay (d_prop)

  The time it takes for the bits of the packet to travel from one end of the link to the other. This delay depends on the physical distance between the two routers and the propagation speed of the medium (e.g., fiber, copper).

Protocol Layers: Structuring Network Communication

Network communication is organized into layers, with each layer implementing a specific service through its internal actions.

Each layer adds its own auxiliary header to the data packet as it passes through, similar to adding addresses to letters, allowing it to be processed correctly by subsequent layers.

Application Layer: Overview

The Application Layer is the top layer, where network applications and their protocols reside.

Client-Server Paradigm

A common architecture where:

• Server: Always-On Host

  The server is a host that is always on and has a permanent IP address (e.g., a DNS server).

• Client: Initiates Communication

  The client initiates communication with the server, is not always on, and typically has a dynamic IP address.

Peer-to-Peer (P2P) Architecture

In P2P applications (like BitTorrent), peers request services directly from other peers, acting as both clients and servers.

Process Communication

A process is a program running within a host.

• Inter-Process Communication

  Within the same host, two processes communicate using inter-process communication, defined by the operating system.

• Cross-Host Communication

  Processes in different hosts communicate by exchanging messages.

• Client Process vs. Server Process

  The client process initiates the communication, while the server process waits to be contacted.

Sockets: The Communication Portal

A socket acts as a portal or door for a process to send and receive messages across the network.