When you send a WhatsApp message, watch a YouTube video, visit a website, make a video call, play an online game, or download a file, information has to travel across a huge network of computers and communication systems before reaching you.
But how does all of that information move from one computer to another?
One of the most important answers is packet switching.
Packet switching is one of the fundamental technologies behind the modern Internet. Instead of sending an entire message through one continuous communication path, packet switching breaks information into smaller pieces called packets. These packets can travel across a network and are then reassembled at their destination.
The idea sounds simple today, but it represented a major change in the way computers communicated.
The history of packet switching goes back to the 1960s, when researchers in the United States and Britain were looking for better ways to move digital information between computers. Their work eventually influenced ARPANET, TCP/IP, and ultimately the global Internet.
This article explains the history of packet switching, the people behind it, why it was invented, how it works, and why it remains essential to the Internet today.
What Is Packet Switching?
Packet switching is a method of transmitting data by dividing a message into smaller units called packets.
Each packet contains a portion of the original information as well as information that helps the network deliver it to the correct destination.
Imagine that you want to send a large book to someone in another city.
Instead of putting the entire book into one enormous package, you could divide it into 100 smaller packages. Each package could contain:
- A portion of the book
- The destination address
- Information identifying the package
- Other information needed for delivery
The packages could travel through different roads and arrive at different times. Once they reach the destination, the recipient puts them back together in the correct order.
That is broadly similar to how packet switching works.
The actual technology is much more sophisticated, but the basic concept is the same.
Packet switching became a foundation of computer networking and was central to the development of ARPANET and the Internet. The Internet Society identifies early packet-switching research as one of the major technological foundations of the Internet.
Why Was Packet Switching Invented?
To understand packet switching, it helps to understand how communication networks worked before it.
For much of the 20th century, telephone communication relied heavily on circuit switching.
When two people made a telephone call, the telephone network established a communication path between them. That path remained allocated for the duration of the call.
This worked well for voice conversations.
However, computers communicated differently from humans.
A computer might send information for a short period, stop, wait, and then send another large amount of information later.
This created an inefficient situation if a communication channel had to remain dedicated even while no data was being transmitted.
Researchers therefore began asking an important question:
Could computers share communication networks more efficiently?
The answer eventually became packet switching.
Circuit Switching vs. Packet Switching
The difference between the two systems can be explained with a simple example.
Circuit Switching
Imagine calling a friend using a traditional telephone system.
A dedicated communication path is established between you and your friend.
For the duration of the conversation, that connection is reserved.
Even when neither person is speaking, the connection remains active.
Packet Switching
Now imagine sending a series of messages through the Internet.
Your information is broken into packets.
The network can use shared communication links to move packets from many different users.
Instead of reserving an entire communication path for one conversation, network resources can be shared among many communications.
This made packet switching particularly attractive for computer networks.
The Early Ideas Behind Packet Switching
Packet switching did not suddenly appear as a single invention by one person.
Several researchers working independently contributed important ideas.
Among the most important names were:
- Leonard Kleinrock
- Paul Baran
- Donald Davies
- Lawrence Roberts
- J.C.R. Licklider
- Robert Kahn
- Vint Cerf
Their work occurred at different organizations and sometimes independently of one another.
The Internet Society notes that research at MIT, RAND, and Britain’s National Physical Laboratory developed along parallel paths during the 1960s.
Leonard Kleinrock and the Mathematical Foundation
One of the earliest important contributors was Leonard Kleinrock, a computer scientist whose research focused on communication networks and queueing theory.
In 1961, Kleinrock published research on the mathematical theory of packet-based communication. He later published a book on the subject in 1964.
His work helped demonstrate that computers could potentially communicate efficiently using packets rather than relying entirely on traditional circuit-switched communication.
The Internet Society identifies Kleinrock’s 1961 work as the first major paper on packet-switching theory.
Kleinrock’s research became particularly important when computer networking projects began moving from theoretical concepts toward real-world systems.
His work at UCLA would later become closely connected with ARPANET.
Paul Baran and Distributed Communications
Another major figure was Paul Baran, a researcher at RAND Corporation.
During the early 1960s, Baran studied how communication systems could continue functioning even if portions of the network were damaged.
This was an important problem during the Cold War.
Traditional communication systems often depended on centralized infrastructure. Baran explored a more distributed approach.
Instead of relying on one central switching point, information could be divided into smaller pieces and sent through a distributed network.
If one part of the network became unavailable, information could potentially use another route.
Baran’s research on distributed communications was published in a series of studies, with major work appearing in the early 1960s. RAND’s historical material describes his work as an important foundation for packet-switching concepts.
An important point, however, is that the modern Internet was not simply created as a network designed to survive a nuclear war.
That popular story is an oversimplification.
Cold War concerns influenced some of the research, particularly Baran’s work on resilient communications, but ARPANET itself also had an important practical goal: allowing researchers to share expensive computing resources and communicate across geographically separated locations.
Donald Davies and the Term “Packet Switching”
Another extremely important contributor was British computer scientist Donald Davies at the National Physical Laboratory (NPL).
Davies independently developed a similar approach to breaking digital information into smaller pieces.
Most importantly, he introduced the term “packet” to describe these pieces of information.
NPL states that Davies and his team developed packet switching during the 1960s, with the technique becoming an important foundation for modern data communication.
Davies recognized that computer communication was different from telephone communication.
Computers often generated information in bursts rather than maintaining a continuous stream.
Breaking information into packets allowed communication infrastructure to be shared more efficiently.
How Packet Switching Works
The basic process can be broken into several steps.
Step 1: Data Is Created
Suppose you send a message through the Internet.
Your device creates digital data representing that message.
The data might be:
- Text
- An image
- Audio
- Video
- A document
- A web request
Step 2: The Data Is Divided
The information is divided into smaller pieces.
These pieces become packets.
A large file might therefore be represented by many packets rather than one enormous block of information.
Step 3: Packets Receive Addressing Information
Packets contain information that helps networking equipment determine where they should go.
Depending on the networking technology and protocol involved, packet headers can contain information such as source and destination addresses and other control information.
Step 4: Packets Travel Through the Network
Routers and other networking devices examine packet information and determine where to forward the packets.
Different packets belonging to the same communication may sometimes take different paths.
Step 5: Packets Arrive at the Destination
The destination computer receives the packets.
Some may arrive in a different order from the way they were sent.
Step 6: The Information Is Reassembled
Protocols and software handle the process of putting the information back together so that the receiving application can use it.
For applications such as web browsing, file transfers, streaming, and messaging, this process happens so quickly that users normally never notice it.
A Simple Real-Life Example
Imagine sending a 100-page document to someone.
Instead of placing all 100 pages into one massive delivery container, the document is divided into smaller packages.
For example:
- Packet 1 contains pages 1–5
- Packet 2 contains pages 6–10
- Packet 3 contains pages 11–15
- Packet 4 contains pages 16–20
And so on.
The packages travel through the delivery system.
One package may take one route while another takes a different route.
When everything arrives, the packages can be organized to reconstruct the document.
The Internet does something conceptually similar with digital information, although modern protocols and networks are far more complicated than this analogy.
The Importance of Routers
Packet switching would not be useful without devices capable of moving packets between networks.
These devices are commonly known today as routers.
Routers examine information associated with packets and use routing information to decide where packets should go next.
In the early ARPANET, an important piece of technology was called the Interface Message Processor, or IMP.
DARPA explains that in 1968 ARPA contracted BBN to build these packet-switching devices, enabling ARPANET to become operational in 1969.
The IMPs played a critical role in connecting the early ARPANET computers.
Packet Switching and ARPANET
Packet switching became particularly important when ARPA began developing ARPANET.
The goal was to connect computers located at different research institutions.
In the late 1960s, Lawrence Roberts led the development of the ARPANET project.
Researchers needed a networking method that could efficiently connect different computers over long distances.
Packet switching provided an attractive solution.
The first four ARPANET nodes were:
- UCLA
- Stanford Research Institute (SRI)
- University of California, Santa Barbara
- University of Utah
The network became operational in 1969. DARPA describes the 1969 four-node network as a major foundation for the Internet that followed.
The First ARPANET Message
On October 29, 1969, researchers at UCLA attempted to send the word “LOGIN” to the Stanford Research Institute.
The system crashed after the first two letters.
The message therefore became famously known as:
“LO”
Although this was a tiny message, the event represented a major milestone.
For the first time, computers at geographically separated research institutions were communicating through a packet-switching network.
The experiment demonstrated that the basic concept could work.
Why Packet Switching Was Better for Computer Networks
Packet switching offered several important advantages.
1. Efficient Use of Network Capacity
Computer data often comes in bursts.
Packet switching allows multiple users to share network infrastructure rather than requiring a dedicated communication circuit for every connection.
2. Better Resilience
Because packets can potentially travel through different paths, a problem affecting one part of a network does not necessarily destroy every communication.
This became particularly important for large distributed networks.
3. Scalability
Packet-based networking made it possible to connect increasing numbers of computers and networks.
As networks expanded, the basic packet concept could continue to be used.
4. Efficient Data Transfer
Large amounts of digital information can be broken into manageable pieces and transmitted through shared infrastructure.
5. Support for Different Applications
The same underlying networking infrastructure can support many kinds of applications.
For example:
- Websites
- Video
- Online gaming
- File transfers
- Voice calls
- Cloud applications
The NPL Packet-Switching Network
The National Physical Laboratory in Britain did not merely study packet switching theoretically.
NPL also built a practical packet-switched network.
NPL’s historical timeline states that by 1970 it had developed a local network using packet switching that supported online services for approximately 200 users.
This was important because it demonstrated that packet switching could move from theory into practical networking.
NPL’s work also influenced the wider development of computer networking.
How ARPANET Helped Prove Packet Switching
ARPANET became one of the most important real-world demonstrations of packet switching.
The network connected computers at research institutions across the United States.
The researchers could use the network to:
- Share computing resources
- Exchange information
- Communicate with other researchers
- Transfer files
- Develop networking software
- Experiment with new communication protocols
The Internet Society notes that ARPANET’s design emerged from work involving researchers at MIT, RAND, NPL, ARPA, UCLA, and other institutions.
The success of ARPANET helped demonstrate that packet-based computer networking could work on a large scale.
From Packet Switching to TCP/IP
Packet switching alone was not enough to create today’s Internet.
There also needed to be a way for different networks to communicate with one another.
This problem eventually led to the development of TCP/IP.
Vint Cerf and Robert Kahn played major roles in developing the architecture that allowed different packet-switched networks to communicate.
This was a crucial step toward the idea of an Internet, meaning a network of interconnected networks.
Instead of having one giant network, different networks could operate independently while using common protocols to exchange information.
That concept became one of the defining characteristics of the modern Internet.
January 1, 1983: TCP/IP Becomes Central
January 1, 1983, is one of the most important dates in Internet history.
ARPANET officially transitioned to TCP/IP.
This allowed different networks to communicate using a common set of protocols.
The change helped establish the foundation for the Internet as a network of networks.
DARPA’s historical timeline identifies 1983 as the year ARPANET switched to TCP/IP.
Packet switching was therefore no longer just an experimental technique.
It had become part of a much larger architecture.
Packet Switching and the Growth of the Internet
During the 1980s and 1990s, computer networking expanded rapidly.
Universities, businesses, governments, and eventually ordinary consumers gained access to increasingly sophisticated networks.
The National Science Foundation’s NSFNET became an important part of the expansion of Internet connectivity in the United States.
Then came the World Wide Web.
In 1989, Tim Berners-Lee proposed the World Wide Web while working at CERN.
The Web made the Internet much easier for ordinary people to use.
People could access information through web browsers rather than needing specialized knowledge of computer networks.
But underneath these websites, packet-based networking was still doing the hard work of moving information.
Packet Switching in the Modern Internet
Today, packet switching is everywhere.
When you open a website, your device exchanges packets with servers.
When you watch a video, the video data travels through networks in packets.
When you send an email, packets carry the information between different systems.
When you play an online game, packets carry information about:
- Player movements
- Game actions
- Positions
- Voice communication
- Server responses
When you use a cloud storage service, packets transport files between your device and remote data centers.
The technology has evolved enormously, but the basic concept remains remarkably important.
Packet Switching and Streaming Video
Consider watching a movie online.
The entire movie is not normally sent to your device as one enormous uninterrupted block.
Digital information is transported through network protocols in smaller units.
Your device receives data, processes it, and continues requesting or receiving additional information.
If your network becomes congested, packets can experience delays or loss, depending on the protocols and application.
Modern streaming systems are designed to deal with these conditions using techniques such as buffering and adaptive bitrate streaming.
This is one reason packet-based networking is so important to modern entertainment.
Packet Switching and Online Gaming
Online games are another excellent example.
Suppose you are playing a multiplayer football game.
Your computer or console needs to communicate information such as:
- Your player’s movement
- Your controller input
- Game events
- Server updates
- Other players’ actions
This information has to move between your device and the game server.
Because the Internet uses packet-based communication, these interactions can occur over shared network infrastructure.
Low latency is particularly important for gaming because delays can affect the player’s experience.
Packet Switching and Mobile Networks
Packet-based networking also became extremely important as mobile communication evolved.
Modern smartphones use packet-based data networks for services such as:
- Web browsing
- Social media
- Video streaming
- Messaging
- Cloud applications
- Online gaming
- Voice-over-IP services
Whether you are using Wi-Fi or cellular data, packet-based communication is a fundamental part of modern digital connectivity.
Packet Switching Made the Internet Scalable
One of the greatest achievements of packet switching is that it helped make large-scale networking possible.
Imagine a network with only a few computers.
Managing communication might be relatively easy.
Now imagine billions of connected devices.
A system requiring a dedicated communication path for every possible interaction would be extremely difficult to manage.
Packet-based networking allows communication resources to be shared across many users and applications.
This flexibility has contributed to the Internet’s extraordinary growth.
The Difference Between Packets and Messages
It is important to understand that a message and a packet are not necessarily the same thing.
A message is the information that a user or application wants to send.
For example:
“Hello, how are you?”
A packet is a smaller unit of data used by a networking system to transport information.
A single message may therefore involve multiple packets.
The networking protocols determine how those packets are addressed, transported, handled, and, when necessary, reconstructed into usable information.
What Happens If a Packet Is Lost?
Networks are not perfect.
Packets can sometimes be lost, delayed, duplicated, or arrive out of order.
This is one reason networking protocols are important.
Some protocols include mechanisms that allow missing information to be detected and retransmitted.
TCP, for example, was designed to provide reliable communication over packet-switched networks.
Other communication methods may prioritize speed over perfect reliability.
This is why different Internet applications can behave differently when network conditions become poor.
Packet Switching and Network Congestion
Imagine a highway during rush hour.
If only a few cars are using the road, traffic moves smoothly.
If thousands of cars enter at the same time, congestion develops.
Computer networks can experience similar problems.
When too much data attempts to travel through limited network capacity, packets may be delayed.
This can result in:
- Slow websites
- Video buffering
- High gaming latency
- Delayed messages
- Poor-quality calls
Modern networking systems use sophisticated techniques to manage congestion and improve performance.
Why Packet Switching Changed the World
Before packet switching, long-distance communication was largely built around systems designed for continuous connections such as telephone calls.
Packet switching introduced a different way of thinking.
Instead of asking:
“How can we create one permanent connection?”
engineers could ask:
“How can we efficiently move small pieces of information across a shared network?”
That change was profound.
It made computer networking more flexible, efficient, distributed, and scalable.
Important People in the History of Packet Switching
Several people deserve recognition for the development of the ideas that eventually became part of Internet networking.
Leonard Kleinrock
Developed important mathematical theories for computer communication and packet switching.
Paul Baran
Worked on distributed communication systems at RAND and explored resilient network designs.
Donald Davies
Independently developed packet-switching concepts at NPL and introduced the term “packet switching.”
Lawrence Roberts
Led the ARPANET project and helped turn networking concepts into a working large-scale network.
J.C.R. Licklider
Promoted the vision of interconnected computers sharing information and resources.
Robert Kahn
Made major contributions to ARPANET architecture and later to internetworking.
Vint Cerf
Co-developed TCP/IP with Robert Kahn, helping establish the technical foundation for connecting different networks.
The development of the Internet was therefore not the work of one person. It was the result of decades of research involving scientists, engineers, universities, government agencies, and private organizations.
Packet Switching Timeline
Here is a simplified timeline of the most important developments.
| Year | Development |
|---|---|
| 1961 | Leonard Kleinrock publishes important research on packet-based communication theory |
| 1964 | Paul Baran’s distributed communications research is published |
| 1965 | Donald Davies develops packet-switching concepts at NPL |
| 1965 | Lawrence Roberts and Thomas Merrill demonstrate a computer-to-computer wide-area connection |
| 1967 | ARPANET plans develop further and packet-switching research from NPL becomes known to ARPA researchers |
| 1968 | ARPA contracts BBN to build Interface Message Processors |
| 1969 | ARPANET becomes operational |
| 1970 | NPL develops a packet-switched local network |
| 1970s | Packet-switching research expands rapidly |
| 1970s | TCP/IP development begins |
| 1983 | ARPANET transitions to TCP/IP |
| 1980s–1990s | Internet networking expands worldwide |
| 1990s | Commercial Internet access and the World Wide Web accelerate adoption |
| Today | Packet switching remains fundamental to Internet communication |
The Internet Society’s historical account confirms the parallel development of packet-switching research at MIT, RAND and NPL, followed by its incorporation into ARPANET.
Was Packet Switching Invented by One Person?
No.
This is one of the most important things to understand about Internet history.
There is no single person who can accurately be described as the sole inventor of packet switching.
Leonard Kleinrock contributed important mathematical theory.
Paul Baran developed distributed communication concepts.
Donald Davies independently developed packet switching at NPL and introduced the terminology.
Other researchers and engineers transformed these concepts into operational networks.
The Internet itself was similarly developed through contributions from many people over several decades.
Packet Switching vs. the Internet
Packet switching and the Internet are not the same thing.
Packet switching is a networking method.
The Internet is a massive global system of interconnected networks that uses packet-based technologies along with many protocols and technologies.
An easy way to understand the relationship is:
Packet switching → ARPANET → TCP/IP → interconnected networks → modern Internet
This is simplified, because the real history contains many parallel developments, but it illustrates the general progression.
Why Packet Switching Still Matters Today
More than half a century after its early development, packet switching remains central to digital communication.
The technology supports an enormous range of activities.
Every day, people use networks for:
- Search engines
- Social media
- Online banking
- E-commerce
- Video calls
- Streaming
- Cloud computing
- Online education
- Remote work
- Gaming
- Artificial intelligence services
- Internet of Things devices
All of these services depend on networks capable of moving digital information efficiently.
The technology behind today’s Internet is vastly more sophisticated than the networks of the 1960s, but the basic idea of dividing and transporting digital information remains fundamental.
Frequently Asked Questions
What is packet switching in simple terms?
Packet switching is a method of sending information by breaking it into smaller pieces called packets. These packets travel through a network and are handled by networking equipment before reaching their destination.
Who invented packet switching?
Packet switching was developed through the work of several researchers. Paul Baran, Donald Davies, and Leonard Kleinrock are among the most important early contributors.
Why was packet switching important?
It allowed computers to share communication infrastructure more efficiently and helped create networks that could grow and connect many different computers.
Did packet switching create the Internet?
Packet switching was one of the fundamental technologies that made the Internet possible, but it was not the only technology involved. ARPANET, TCP/IP, routing, networking hardware, DNS, the World Wide Web, and many other developments also played important roles.
What is the difference between packet switching and circuit switching?
Circuit switching establishes a dedicated communication path, while packet switching divides information into packets that can share network resources.
Is packet switching still used today?
Yes. Packet-based communication remains fundamental to modern computer networking and the Internet.
What was ARPANET's role in packet switching?
ARPANET was one of the earliest major computer networks to demonstrate packet-switching technology on a practical scale. It became an important predecessor to the modern Internet.
Conclusion
The history of packet switching is one of the most important chapters in the history of the Internet.
The idea emerged during a period when computers were becoming increasingly powerful but communication systems were still largely designed around traditional telephone technology.
Researchers such as Leonard Kleinrock, Paul Baran, and Donald Davies explored new ways to move computer information through networks.
Their work helped establish the concept of breaking information into packets and moving those packets through shared communication networks.
The ideas eventually influenced the development of ARPANET.
ARPANET demonstrated that geographically separated computers could communicate through a packet-switched network. Later, TCP/IP provided a way for different networks to communicate with one another.
From there, the Internet continued to expand.
Today, when you stream a movie, send a message, visit a website, use social media, play an online game, or access a cloud application, you are benefiting from the networking principles developed decades ago.
Packet switching may sound like a technical term from computer history, but it is actually one of the technologies that quietly powers everyday digital life.
The next major chapter in this history is the story of the first Internet message in 1969—the famous “LO” that was sent between UCLA and the Stanford Research Institute during the first ARPANET connection.
