How the Internet Works: Packets, Protocols, and Global Infrastructure

What Is the Internet?

The internet is not a single network but a global network of networks — billions of computers and devices connected through shared protocols that enable them to communicate regardless of who manufactured them, what operating system they run, or where in the world they are located. The genius of the internet's design is that it defines only the rules for communication (the protocols), not the physical infrastructure — allowing extraordinary diversity of hardware and technology underneath.

The internet was not planned as a civilian utility. Its origins lie in ARPANET, a US Defense Department research network created in 1969 to connect university research computers. The first message sent over ARPANET ("login") crashed the system after two characters. From that modest beginning grew a network now connecting approximately 5.5 billion people.

Packet Switching: How Data Travels

The internet moves data through packet switching — a radically different approach from the dedicated circuits of telephone networks. When you send data (a webpage request, email, video stream), it is broken into small chunks called packets — typically 1,000–1,500 bytes each. Each packet contains:

• The destination IP address
• The source IP address
• A sequence number (to reassemble in order)
• The data payload
• Error-checking information

Packets from the same message may travel different routes across the network, each independently finding its way to the destination, where they are reassembled in order. This makes the network resilient — if one path is congested or broken, packets route around it. It also allows the network's capacity to be shared efficiently among many simultaneous users.

TCP/IP: The Internet's Languages

The protocols that enable internet communication are collectively called TCP/IP — the Transmission Control Protocol and Internet Protocol. They were designed by Vint Cerf and Bob Kahn in the early 1970s and adopted as the internet's standard protocols in 1983.

Internet Protocol (IP) handles addressing and routing — defining how packets are labeled with source and destination addresses and how they navigate from router to router across networks. Every device on the internet has an IP address — a unique numerical identifier (IPv4 addresses like 192.168.1.1; IPv6 addresses accommodate far more devices).

Transmission Control Protocol (TCP) handles reliable delivery — ensuring that all packets arrive, in order, without errors. TCP establishes connections, handles retransmission of lost packets, manages flow control (preventing fast senders from overwhelming slow receivers), and confirms delivery. For applications where some packet loss is acceptable (video calls, live streaming), UDP (User Datagram Protocol) is used instead — faster but without TCP's reliability guarantees.

These protocols form the foundation of a layered network model. Above TCP/IP sit application protocols: HTTP/HTTPS (web), SMTP (email), FTP (file transfer), DNS (domain name resolution), and many others.

DNS: The Internet's Phone Book

Humans use domain names (google.com, wikipedia.org) while the internet routes by IP addresses. The Domain Name System (DNS) translates between them. When you type a URL, your computer queries DNS servers in a hierarchical process:

1. Your computer checks its local cache — if it recently looked up this domain, it uses the cached answer
2. If not cached, it queries your ISP's DNS resolver
3. If the resolver doesn't know, it queries a root DNS server (there are 13 root server clusters worldwide)
4. The root server directs the query to the appropriate top-level domain (TLD) server (.com, .org, etc.)
5. The TLD server directs to the authoritative nameserver for the specific domain
6. The authoritative nameserver returns the IP address

The entire process typically completes in milliseconds. The result is cached at each level to avoid repeating the lookup for the same domain.

What Happens When You Load a Webpage

When you type a URL and press Enter, a complex sequence occurs in fractions of a second:

1. DNS resolution: Your browser resolves the domain name to an IP address
2. TCP connection: Your browser establishes a TCP connection to the server (a three-way handshake: SYN, SYN-ACK, ACK)
3. TLS handshake: For HTTPS, an additional cryptographic handshake establishes an encrypted connection (using TLS/SSL)
4. HTTP request: Your browser sends an HTTP GET request: "Give me this URL"
5. Server processing: The web server receives the request, retrieves or generates the content, and sends back an HTTP response with the HTML, images, and other resources
6. Rendering: Your browser parses HTML, requests additional resources (CSS, JavaScript, images), and renders the page

Physical Infrastructure

The internet's physical substrate is extensive and diverse:

• Submarine cables: Approximately 400 submarine fiber optic cables carry ~95% of international internet traffic along the ocean floor. The transatlantic cable carries millions of gigabits per second. These cables are surprisingly vulnerable — ships dragging anchors are a common cause of cable breaks.
• Internet exchange points (IXPs): Physical locations where different networks interconnect, exchanging traffic without paying transit costs. The largest IXPs in Frankfurt, Amsterdam, and London handle petabits per second.
• Data centers: Massive facilities housing the servers of cloud providers, websites, and streaming services. A large hyperscale data center may consume 100+ megawatts of power — more than a small city.
• The "last mile": The connection from the internet backbone to your home or office — via fiber optic cable, coaxial cable (cable internet), copper telephone lines (DSL), or wireless (4G/5G, satellite). This is often the bottleneck for household internet speeds.