A Web Address Finds a Slice of a Shared Computer

The page that appears when you type a website name does not come from a single box with that name on it. It comes from a slice of a very large computer, found by looking up a number.

In modern computing, work rarely happens directly on bare hardware. Instead, operators use a virtual machine, which is software that simulates hardware, to divide one powerful computer into several smaller simulated ones. Those systems support the wider web.

Each machine is connected under the internet protocol and has a unique IP address that identifies it on the network, much like a street address. That address is usually hidden behind a URL, the familiar web address, which is listed in a worldwide directory known as the domain name service. Entering the URL prompts a lookup that returns the matching address.

Then the two computers open a connection with a TCP handshake, a short introductory exchange that lets them swap data bursts known as packets.

Above that connection, there is often a protective setup, for example SSL, for scrambling messages when they leave one machine and unscrambling them when they reach the other.

Then the two machines can exchange information using the hypertext transfer protocol. Your computer, acting as the client, may request a web page, and the server responds with the code for the page. Modern servers also offer a common method letting a client ask for data, known as an API, short for application programming interface.

Then the page appears on your screen.