Computer Networking — Chapter 1: Computer Networks and the Internet

What Is the Internet?

The Internet can be described in two ways:

Nuts-and-Bolts Description — The Internet is a computer network that interconnects billions of computing devices (called hosts or end systems) worldwide through a network of communication links and packet switches (routers and link-layer switches). End systems access the Internet through Internet Service Providers (ISPs).

Services Description — The Internet is an infrastructure that provides services to distributed applications (e.g., Web, email, streaming, social networks). Applications run on end systems, not on packet switches. End systems provide a socket interface that specifies how a program asks the Internet to deliver data to a destination program.

Definition of a Protocol. A protocol defines the format and the order of messages exchanged between two or more communicating entities, as well as the actions taken on the transmission and/or receipt of a message or other event.

Internet standards are developed by the Internet Engineering Task Force (IETF). Standards documents are called Requests for Comments (RFCs).

The Network Edge

End systems are also referred to as hosts. Hosts are divided into clients and servers. Most servers reside in large data centers (e.g., Google has 50-100 data centers).

Access Networks — The network that physically connects an end system to the first router (edge router):

Physical Media — Two categories:

• Guided media: twisted-pair copper wire, coaxial cable, fiber-optic cable
• Unguided media: terrestrial radio spectrum, satellite radio spectrum

Optical Carrier (OC) standard link speeds: OC-n = n × 51.8 Mbps (OC-1, OC-3, …, OC-768).

The Network Core

Packet Switching — End systems break data into packets. Each packet travels through the network independently. Store-and-forward transmission: a packet switch must receive the entire packet before it can begin to transmit on the output link.

Store-and-Forward Delay. If a packet of L bits is transmitted at rate R bps, the time to push the packet onto the link is L/R seconds. With N links of rate R between source and destination, the end-to-end delay for one packet is: d_end = N × (L/R)

Circuit Switching — A dedicated end-to-end circuit is established before communication. Resources along the path are reserved for the duration. Example: traditional telephone network. Uses Frequency-Division Multiplexing (FDM) or Time-Division Multiplexing (TDM).

Key Comparison:

• Packet switching: no reservation, statistical sharing, can be “best-effort”
• Circuit switching: guaranteed resources, but less efficient for bursty traffic

A Network of Networks — ISPs are interconnected in a hierarchy:

• Access ISPs (residential, corporate, university)
• Regional ISPs
• Tier-1 ISPs (e.g., Level 3, AT&T, Sprint, NTT)
• IXP (Internet Exchange Point) — a meeting point where multiple ISPs peer together
• PoP (Point of Presence) — a group of routers in an ISP’s network where routers from other ISPs can connect

Delay, Loss, and Throughput

Four Sources of Packet Delay:

1. Processing delay (d_proc) — time to examine packet header, check bit-level errors
2. Queuing delay (d_queue) — time waiting in output buffer for transmission
3. Transmission delay (d_trans = L/R) — time to push all bits of packet onto the link
4. Propagation delay (d_prop = d/s) — time for signal to propagate across the medium

Total Nodal Delay: d_nodal = d_proc + d_queue + d_trans + d_prop

Traffic Intensity — Let L = packet length (bits), a = average packet arrival rate (packets/sec), R = transmission rate (bps). Traffic intensity = La/R.

• If La/R > 1: queue grows without bound (infinite delay)
• If La/R <= 1: queuing delay depends on burstiness of arrivals

Packet Loss — When a queue (output buffer) is full, arriving packets are dropped (lost).

Throughput — The rate at which bits are transferred from sender to receiver. For a path with multiple links, the bottleneck link (the link with the smallest transmission rate) determines the end-to-end throughput.

Protocol Layers and Service Models

Five-Layer Internet Protocol Stack:

OSI Model adds two layers: Presentation and Session.

Encapsulation — Each layer takes data from the layer above, adds header information, and passes it to the layer below. At the receiving end, the process is reversed (de-encapsulation).

Networks Under Attack

• Malware: viruses (user interaction needed) and worms (self-propagating without user action)
• Botnet: network of compromised devices controlled by attackers
• DoS (Denial-of-Service) attacks:
  • Vulnerability attack (crash the target with crafted messages)
  • Bandwidth flooding (overwhelm the target’s access link)
  • Connection flooding (bogus TCP connections)
• Packet sniffing: passive receiver captures packets (e.g., WiFi)
• IP spoofing: sending packets with a false source address

Key Formulas