How the Domain Name System Works

From ICANNWiki
Jump to navigation Jump to search

The Domain Name System (DNS) is the system used to translate alphanumeric domain names into Internet Protocol numbers. Simply put, the DNS converts the names typed in the Web browser address bar into IP addresses. [1]


The DNS is made up of many servers and databases which, through a series of lookups in various caches, configure Domain Names into IP Addresses. The Domain Name System is a distributed database arranged hierarchically; its purpose is to provide a layer of abstraction between Internet services (web, email, etc.) and the numeric addresses (IP addresses) used to uniquely identify any given machine on the Internet. The DNS associates a variety of information with the domain names assigned and, most importantly, translates the domain names meaningful to humans into the numerical identifiers that locate the desired destination.

PowerDNS has a primer on the DNS as it functioned in 2018: A warm welcome to DNS


The need for a hierarchical DNS arose with the popularity of the Internet in academic spheres in the early 1980s, which eventually necessitated a decentralized Internet. Communications between The Stanford Research Institute NIC and other institutions included plans to create a hierarchical DNS, and can be found in RFC 805, a group document from 1982. This document outlines many of the basics of the eventual DNS, including the need for TLDs to provide a fixed starting point for queries, and the need for SLDs to be unique. This, in turn, would necessitate the need for a registrar type of administration, and help the nascent IT community recognize that the distribution of responsibility for each domain to individual name servers would provide administrative advantages.[2]

How Does It Work?

The DNS makes it possible to assign domain names in a meaningful way to Internet resources as well as to users, regardless of the entity's location. As a result, the WWW hyperlinks remain consistent, even for mobile devices. A domain name is an easy way to remember an address, but that needs to be converted to its numerical, IP format. [3]

Coordination across the Internet is maintained by means of a complex authoritative root system known as the Top Level Domain (TLD), as well as the DNS and other smaller name servers responsible for hosting individual domain information.

DNS includes three types of top-level domains: generic (gTLD), country code (ccTLD), and sponsored (sTLD). gTLDs include domains that could be obtained by anyone (.com, .info, .net, and .org). Since 2014 many other gTLDs have been added like .pub, .ngo, .sucks. sTLDs are limited to a specific group e.g .aero (for air-transport industry).

For each domain, the DNS spreads the responsibility by mapping the domain names and assigning them into IP addresses, and vice-versa. This is accomplished through authoritative name servers which have been designated for each domain. Each authoritative name server is responsible for its own particular domain, but it has the authority to assign new authoritative name servers to any of its sub-domains. The DNS is able to store many types of information, even the mail server lists for a specific domain. The DNS is a core element which ensures the functionality of the Internet through its distributed keyword-based redirection service.

However, the DNS does not include security extensions, which was instead developed as DNSSEC.

The Structure of a DNS

The Domain Name System presents the following structure:

  • Domain space name: represented by tree of domain names with nodes and leaves [4]
  • Domain name syntax: rules include in standards like RFC 1035, RFC 1123, and RFC 2181
  • Name server
  • Domain names Internationalized
  • DNS resolver: initiates the queries will finally lead to the complete translation (resolution) of the information.

DNS Services

These types of services include:

  1. Managing the DNS
    1. By means of powerful, secure and complete tools and DNS administration options such as Manage A records or MX records
    2. By means of controlling and managing the traffic
  2. Hosting the DNS
    1. Efficient DNS resolution
  3. Location DNS services: to ease visitors' work, increase visitors' satisfaction and strengthen the relationship with all visitors.[5]

Developments and the "DNS Camel"

In 2012, an IETF draft proposal, called "DNS Extension for Autonomous Internet (AIP)," was written by three Chinese technologists suggesting a method of operating alternate DNS root servers within national boundaries using gateways for translation. All DNS requests would carry an additional TLD, in order to designate that the requests were being sent to an alternate root. The standard would add a single letter to the DNS string - for example, "" - to designate the address of the alternate, nationally-controlled root.[6]

At the time, critics noted that the proposal would allow greater governmental control over the Internet.[7] The Internet Draft expired in April 2018 after eleven proposed revisions.[6] The draft was one of many efforts to create alternative root systems for the Internet.

The DNS Camel refers to the overloading of features and protocols onto the DNS architecture, such that one small addition will "break the camel's back."


Related Articles