Internet Protocol

The Internet Protocol, usually called IP, is the fundamental set of rules that governs how data is addressed and routed across the internet and other computer networks. Every device connected to the internet, from a smartphone to a massive data center server, is assigned an IP address, a unique numerical label that identifies it on the network, similar to how a postal address identifies a specific building. IP defines how data is broken into small chunks called packets, how each packet is labeled with its destination address, and how networking equipment along the way reads that address to decide where to send the packet next, without needing to understand anything about the content inside it.

Packets and Addressing

Rather than sending an entire file or message as one continuous stream, IP breaks data into small, independently addressed packets, each carrying a header with the sender’s and receiver’s IP addresses along with the information needed to reassemble the full message once every packet arrives. This packet-based design, called packet switching, lets a single network connection carry traffic from many different conversations at once and lets individual packets travel by different paths and arrive out of order, only to be reassembled correctly at the destination, which makes the overall network far more resilient and efficient than reserving a dedicated connection for every single conversation.

IP addresses come in two current versions. IPv4, the original version still widely used today, represents each address as four numbers separated by dots, such as 192.0.2.1, which allows for about 4.3 billion unique addresses, a number that has proven too small for a world with many billions of connected devices. IPv6, developed specifically to solve this shortage, uses a much longer address format capable of generating an astronomically larger number of unique addresses, enough to assign a vast number of unique addresses to every person on Earth many times over, and adoption of IPv6 has grown steadily as the pool of available IPv4 addresses has become increasingly scarce.

IP’s Place in the Larger System

IP is deliberately a simple, best-effort protocol: it tries to deliver each packet but makes no guarantee that a packet will actually arrive, arrive only once, or arrive in the correct order. Reliability is handled by a companion protocol, the Transmission Control Protocol, or TCP, which runs on top of IP, checking that every packet arrives, requesting that any missing or corrupted packets be resent, and reassembling packets in their correct order before handing a complete, reliable stream of data to the application using it, such as a web browser loading a page. This pairing is so fundamental to how the internet works that the combined system is often referred to simply as TCP/IP.

Devices called routers sit at the junctions of different networks and use IP addresses to decide how to forward each packet one step closer to its destination, consulting routing tables that describe which direction leads toward which range of addresses. Because no single router needs to know the full path a packet will take, only the next step, this distributed approach lets the internet scale to billions of devices without requiring any single point of centralized control, and it is a major reason the internet has continued to function reliably even as it has grown far beyond what its original designers imagined in the 1970s and 1980s.

Origins and Ongoing Role

The Internet Protocol grew out of research funded by the United States Defense Advanced Research Projects Agency in the 1970s, aimed at connecting different, independently designed computer networks into a single unified system, a goal that led directly to the TCP/IP design still in use today. The protocol was formally standardized and the early ARPANET research network switched over to it on January 1, 1983, a date often treated as a symbolic birth of the modern internet, since it marked the point at which a common addressing and routing scheme let previously separate networks begin to interconnect freely.

Despite the internet growing from a small research network into a system connecting billions of devices worldwide, the core design principles behind IP, simple best-effort packet delivery, distributed routing decisions, and a layered relationship with protocols like TCP that add reliability on top, have remained remarkably stable for decades. Most of the major changes since then, including the long transition toward IPv6 and the rise of techniques for managing scarce IPv4 addresses, have extended the original design rather than replacing it, a testament to how well the protocol’s original, deliberately simple architecture has scaled.

Frequently Asked Questions

What is the difference between IPv4 and IPv6? IPv4 uses a shorter address format that allows for about 4.3 billion unique addresses, which has become insufficient given the number of connected devices worldwide. IPv6 uses a much longer address format that provides a vastly larger pool of addresses, and the two coexist today while adoption of IPv6 continues to grow.

Why is IP described as best-effort rather than reliable? IP itself only tries to deliver packets without guaranteeing they will arrive, arrive once, or arrive in order, keeping the protocol simple and efficient. Reliability is instead provided by a separate protocol, typically TCP, that runs on top of IP and handles retransmission and correct ordering.

How does a router know where to send a packet? A router reads the destination IP address in a packet’s header and consults a routing table that maps ranges of addresses to the next network hop, forwarding the packet one step closer to its destination without needing to know the complete end-to-end path in advance.

Editorially reviewed by Specialty Digest Editorial TeamLast reviewed November 2, 2026Researched and drafted with AI assistanceReport an issue