Packet Switching

Packet switching is a method of data transmission in which messages are broken into discrete blocks — packets — that travel independently across a network and are reassembled at the destination. Unlike circuit switching, which reserves a dedicated communication path for the duration of a session, packet switching shares network resources dynamically, making it radically more efficient for bursty data traffic and fundamentally more resilient to partial network failure.

The concept was developed independently by Paul Baran at RAND Corporation and Donald Davies at the National Physical Laboratory in the 1960s, though its philosophical roots trace back to the realization that communication networks could function like postal systems rather like telephone exchanges. The ARPANET was the first large-scale implementation, and packet switching remains the foundational architecture of the Internet, Ethernet, and virtually all modern data networks.

Packet switching was not merely an engineering improvement over circuit switching; it was a reconceptualization of what a network is. Where circuit switching treats the network as a pipe, packet switching treats it as a shared resource that can be dynamically allocated. This shift from dedicated to statistical multiplexing is the single most important architectural decision in the history of networking, and it is still poorly understood by policymakers who regulate networks as if they were pipelines.