Chapter 1 Review

The map

The recap

What the Internet is made of

The chapter opened by answering what is the Internet? twice — as hardware and software, and as a service for distributed applications. Both answers are needed, and the second is why the book is organised top-down.

Then it worked inward. At the network edge are the hosts host Simple Any device connected to the Internet that runs your programs: a laptop, a phone, a server. Precise A device connected to the Internet that runs (hosts) application programs. Also called an end system, because it sits at the edge of the network; the book uses host = end system. introduced in ch. 1 — open in glossary that run the applications. Reaching them are access networks — DSL (Digital Subscriber Line) Digital Subscriber Line Home Internet access over the existing telephone wire. introduced in ch. 1 , cable, fibre to the home, 5G fixed wireless, Ethernet and WiFi (IEEE 802.11 wireless LAN) IEEE 802.11 wireless LAN Wireless LAN access; a user must be within a few tens of metres of the access point. WiFi is a trade name, not an abbreviation. introduced in ch. 1 — and beneath those, the physical media that actually carry the bits.

How data actually moves

At the network core the book set out the chapter’s sharpest contrast.

Packet switching packet switching Simple Sending data as separate packets that share links with everyone else's packets, on demand. Precise The approach in which a source breaks a long message into packets that travel through links and packet switches without reserving any link resources. Link capacity is allocated on demand, packet by packet, only among users who have packets to send. introduced in ch. 1 — open in glossary sends packets with no reservation, and shares each link on demand. Circuit switching reserves a share of every link in advance, using FDM (Frequency-Division Multiplexing) Frequency-Division Multiplexing Gives each connection its own frequency band on the link, all of the time. introduced in ch. 1 or TDM (Time-Division Multiplexing) Time-Division Multiplexing Gives each connection the whole link, but only during its own repeating time slot. introduced in ch. 1 , and guarantees a constant rate. Packet switching won because user traffic is bursty: a link that supports 10 reserved users supports 35 on-demand users at essentially the same quality.

The price is that packets wait. Store-and-forward means every switch receives a whole packet before sending any of it, so one packet over $N$ links costs $N \times L/R$ — the chapter’s first equation.

How well it performs

The second half of the chapter stopped describing and started measuring.

Four delays at every node — processing, queuing, transmission, propagation — add up to the nodal delay nodal delay Simple The total time a packet spends at one router. Precise d_nodal = d_proc + d_queue + d_trans + d_prop: the sum of the four delay components suffered at a single node. introduced in ch. 1 — open in glossary and accumulate into the end-to-end delay. The pair that must never be confused is $d_{\text{trans}} = L/R$, which depends on packet length and link rate, and $d_{\text{prop}} = d/s$, which depends on distance.

Queuing delay is the only one that varies from packet to packet. It is governed by the traffic intensity $La/R$, which explodes as it approaches 1. Real buffers are finite, so instead of infinite delay you get packet loss — silently, with nothing reported to the sender.

Throughput throughput Simple How many bits per second actually arrive at the receiver. Precise The rate, in bits/second, at which the receiving host receives data. Instantaneous throughput is that rate at an instant; the average throughput of a transfer of F bits taking T seconds is F/T. With no competing traffic it is min{R₁, …, R_N}. introduced in ch. 1 — open in glossary is set by the bottleneck link: $\min$ of the rates along the path. In today’s Internet that is usually the access network, because the core is over-provisioned. The exception is a core link shared by many flows, where what matters is $R/M$ rather than $R$.

How it is organised

Two organising ideas, and they are the ones later chapters lean on hardest.

The Internet is a network of networks network of networks Simple The Internet is not one network but many separate networks joined together. Precise The structure of today's Internet: hundreds of thousands of access ISPs interconnected through regional ISPs and about a dozen tier-1 ISPs, together with PoPs, multi-homing, peering, IXPs and content-provider networks. introduced in ch. 1 — open in glossary : hundreds of thousands of independently managed ISPs, arranged in a hierarchy that economics built. Access ISPs pay regional ISPs, regional ISPs pay tier-1 ISPs, and tier-1 ISPs pay nobody. Wherever two networks can avoid paying a third, they connect directly.

The Internet has five layers, and each wraps what it is handed in its own header.

And what goes wrong

Malware, botnets, three kinds of denial-of-service attack, passive sniffing and address spoofing — and behind all of them one architectural fact. The Internet was designed for a group of mutually trusting users attached to a transparent network, a model in which security is unnecessary. That assumption no longer holds, and chapter 8 is the repair job.

The road ahead

The book’s remaining chapters are organised around the layers, taken top-down.

Check yourself