Circuit Switching vs. Packet Switching

Circuit Switching: The Dedicated Path

Before the internet, the dominant model for network communication was circuit switching. The classic example is the traditional telephone system. When you made a call, the network established a dedicated, physical connection—a "circuit"—between you and the person you were calling. This entire circuit was reserved for your conversation for its entire duration, whether you were speaking or not.

To get a clear, animated explanation of how this works, please watch the following short video clip.

As the video showed, circuit switching has a few defining traits:

• Dedicated Path: A single, end-to-end path is established and reserved for one communication session.
• Guaranteed Resources: Because the path is dedicated, you have a guaranteed amount of bandwidth. This results in consistent quality and very low, predictable latency, which is excellent for real-time voice calls.
• Inefficiency: The biggest drawback is that these dedicated resources are wasted. During the pauses in a phone conversation, the line sits idle, but no one else can use it. Computer data is often "bursty"—sent in short, intense bursts with periods of silence. Dedicating a full-time circuit for this is highly inefficient and expensive.

The limitations of circuit switching for computer communication were a major hurdle. A new method was needed that could handle bursty data traffic efficiently and be more resilient to network failures.

Packet Switching: The Revolutionary Idea

The solution came from a radically different way of thinking about data. Instead of sending a continuous stream of information over a dedicated line, what if you broke the data into small, discrete chunks?

This is the core idea of packet switching. A message (like an email, an image, or a part of a webpage) is divided into smaller pieces called packets. Each packet is like a small envelope containing a piece of the data, along with a "mailing label" (the header) that has the destination address.

These packets are then sent out into the network independently. They can travel along different paths, get routed around congested areas or broken links, and arrive at the destination out of order. The receiving computer then uses the information in the packet headers to reassemble them into the original message.

Let's watch another segment from the same animated video to see this in action.

This approach directly addressed the problems of circuit switching.

• Efficiency: Network lines are shared. Packets from many different users can be interleaved on the same line, making much better use of available bandwidth. There are no "idle" lines.
• Resilience: If one path fails, routers can simply send the packets along a different route. The network can survive the loss of some of its links, which was a key requirement for the ARPANET project.

This idea wasn't just a technical tweak; it was a fundamental paradigm shift driven by the need for a "survivable communication network," as pioneer Paul Baran called it. To understand the historical and conceptual origins of this idea, the following video provides excellent context.

Head-to-Head Comparison

Now that we've defined both methods, let's put them side-by-side. The best way to grasp the differences is to look at a structured comparison.

As you can see from the table, there's a clear trade-off. Circuit switching offers predictable performance at the cost of inefficiency. Packet switching offers efficiency and resilience at the cost of some unpredictability (variable latency, the need for more complex protocols).

For the bursty, diverse, and unpredictable nature of computer data, the choice was clear. The ARPANET, and the internet that grew from it, was built on packet switching.

This choice had profound implications. By creating a simple, flexible network that didn't make assumptions about what it would be used for, the internet's designers built a system that could adapt to anything. It was, as the following clip explains, "good enough for everything, but optimal for nothing."

This principle is a powerful lesson in system design that resonates strongly with computer science: designing for flexibility often yields a more robust and long-lived system than designing for narrow, short-term performance goals.

Conclusion

In this lesson, we dissected the two fundamental approaches to network switching.

Key Takeaways:

• Circuit Switching creates a dedicated, end-to-end circuit for a communication session. It guarantees performance but is inefficient for bursty computer data. It was the model of the old telephone network.
• Packet Switching breaks data into small packets that travel independently over a shared network and are reassembled at the destination. It is highly efficient and resilient.
• The internet is built on packet switching because its flexibility, efficiency, and robustness were perfectly suited to the needs of computer networking. This design choice allowed it to evolve and support an incredible variety of applications over time.

Preview of the Next Lesson:

Now that we understand the technological foundation of packet switching, our next lesson will focus on its first major implementation: the ARPANET. We will explore its creation, its purpose, and how it laid the groundwork for the global internet we use today.