Circuit switching versus packet switching

As explained earlier, circuit-switching serves as an infrastructure for connection-oriented protocols, while packet-switching serves the connectionless protocols. Both methods deal with data transmission. In circuit-switching, when two devices want to communicate, the connection is established to serve the purpose. As long as the transmission is active, the circuit is only available to the communicating devices. Once the transmission is complete, the connection is broken, and the circuit becomes available for other devices to use. The circuit-switching method is traditionally used for voice communication, and thus the best example is the Public Switched Telephone Network (PSTN), which is still in use worldwide.

Unlike circuit-switching, packet-switching employs a different approach. When two devices want to communicate, no connection is established. Instead, on the sending device, the data is broken into chunks of data known as packets. Each packet contains a header. The header contains a destination IP address and the source IP address. In addition, the header contains the number of the packet and the information into how many packets the data has been broken into. All of that information in the header of a packet is handled by the router, which takes care to send packets on their way. Presumably, packets will travel through many routers, each taking its own path. At the destination, based on the number of the packets located in the header, packets are arranged in the right order. The best example of packet-switching is a computer network.