How your computer reaches
www.bbc.co.uk

The BBC runs its website from data centres — buildings filled with thousands of servers, connected to the internet via multiple high-speed fibre links. To handle millions of simultaneous visitors, traffic is spread across many servers using a load balancer, which routes each incoming request to a server with spare capacity.

For static assets (images, CSS, JavaScript), the BBC uses a CDN (Content Delivery Network) — a global network of cache servers placed close to users. When you request an image, it comes from a CDN node in London rather than from a distant data centre, cutting latency significantly.

The server that responds to your request processes it, retrieves the relevant content from a database if needed, and assembles an HTTP response containing the HTML, which travels back across the network to your browser.

The TCP/IP model has four layers. When you send a request, data is encapsulated — each layer wraps the data in its own header before passing it down. At the destination, layers are unwrapped in reverse, each reading only its own header.

The Application layer contains the actual content — your HTTP request. TCP (Transport layer) breaks this into segments, numbers them, and guarantees they all arrive and are reassembled in order. If a segment is lost, TCP requests it again. IP (Internet layer) assigns source and destination addresses and decides the route. The Network Access layer handles the physical transmission — over Wi-Fi radio waves or Ethernet electrical signals.

Before any data flows, TCP performs a three-way handshake: your computer sends SYN, the server replies SYN-ACK, you confirm with ACK. Only then does the actual request travel.

HTTP (HyperText Transfer Protocol) is the language browsers and servers use. A request specifies a method (GET to fetch, POST to submit data), a path, and headers with metadata. The server replies with a status code (200 OK, 404 Not Found, 301 Redirect) and the response body.

HTTPS is HTTP over an encrypted TLS (Transport Layer Security) connection. Before the HTTP request travels, TLS performs a handshake to agree on encryption and authenticate the server's identity using a certificate issued by a trusted Certificate Authority.

Each packet is typically up to 1,500 bytes. A packet has two parts: a header (source IP, destination IP, sequence number, TTL) and a payload (the actual data chunk). The BBC's 89 kB HTML response is split into roughly 60 packets.

Unlike circuit switching (where old telephone calls reserved a dedicated path end-to-end), packet switching lets each packet take any available route through the network. Routers at each hop read the destination IP and forward the packet toward its destination — different packets from the same page may travel via different cities or even countries.

At your computer, TCP collects all arriving packets, uses the sequence numbers to reorder them correctly (even if they arrived out of order), and requests any that went missing. Only once all packets are accounted for does your browser receive the complete page.