Humans prefer names; Computers prefer numbers.
When you want to visit Google, you type google.com into your browser. You don't type 142.250.190.78. Computers, however, need that IP address to route packets across the internet. The Domain Name System (DNS) is the essential infrastructure that bridges this gap — it translates human-readable names into machine-usable numbers.
DNS is often described as the "phonebook of the internet." You look up a name you recognize (like "John Smith"), and the phonebook gives you the number you need to connect. Without DNS, every user would need to memorize a unique string of numbers for every website, API endpoint, or mail server they wanted to reach. The modern internet would be practically unusable.
DNS is not a single server or a single database. It is a globally distributed, hierarchical system made up of millions of servers working together. No single machine holds all the answers. Instead, responsibility is delegated in a tree structure: a small set of root servers at the top know who is responsible for .com, .org, .io, and every other top-level domain. Those servers delegate to authoritative name servers for each specific domain. This design means the system scales to handle trillions of queries per day without a central bottleneck.
Interviewers often ask: "What happens when you type a URL into a browser?" DNS resolution is the first step. A strong answer covers: browser cache → OS cache → recursive resolver → root server → TLD server → authoritative server → response cached with TTL. Knowing this full chain — not just "DNS looks up the IP" — signals senior-level understanding.