Circuit Switching vs Packet Switching | Generated by AI
Question: How do traditional telephone networks work, and why couldn’t the internet be built directly on top of that infrastructure using its switching method?
Answer:
How Traditional Telephone Networks Work
Traditional telephone networks, known as the Public Switched Telephone Network (PSTN) or Plain Old Telephone Service (POTS), operate using circuit switching.
- When a call is initiated, the network establishes a dedicated physical path (circuit) between the caller and recipient through a series of centralized switching centers (also called central offices or exchanges).
- This path is reserved exclusively for the duration of the call, even during silences, ensuring constant bandwidth and low latency for real-time voice transmission.
- The infrastructure primarily consists of copper wires, fiber optics, trunk lines, and switching nodes that route the analog (or later digital) voice signals.
- Once the call ends, the circuit is released.
This design made the network reliable for voice but vulnerable: destroying key switching centers could isolate large areas or cripple the system, as there was no automatic rerouting around failures.
Why the Internet Couldn’t Be Built Directly on Top of Circuit-Switched Telephone Networks
The internet was designed for data communication (not real-time voice), requiring different priorities: efficiency, resilience, scalability, and support for bursty traffic (e.g., file transfers, emails). Building it on circuit switching would have been impractical for several reasons:
- Inefficiency for Data Traffic:
- Data transmission is often “bursty” — large amounts sent intermittently, with long idle periods.
- Circuit switching wastes bandwidth by holding a dedicated path open during idle times, making it costly and inefficient for non-continuous data.
- Lack of Resilience:
- Circuit-switched networks rely on centralized switches; failure of a key node disrupts connections with no automatic recovery.
- Early internet designers (influenced by Cold War needs via ARPANET) prioritized survivability: the network had to route around damage (e.g., nuclear attack). Packet switching allows this naturally.
- Poor Scalability and Resource Sharing:
- Dedicated circuits limit simultaneous connections and don’t allow multiple users to efficiently share bandwidth.
- Data networks need to connect diverse computers with varying speeds and protocols, which circuit switching couldn’t handle flexibly.
Instead, the internet uses packet switching (pioneered in ARPANET, 1969):
- Data is broken into small packets, each with header information (source, destination, sequence).
- Packets travel independently over shared links, routed dynamically by routers.
- They can take different paths, be reassembled at the destination, and rerouted if links fail.
- This enables efficient bandwidth use, fault tolerance, and scalability.
Historical Context: ARPANET chose packet switching over circuit switching for resource sharing among computers, cost-effectiveness, and robustness (ideas from Paul Baran, Donald Davies, and others). While dial-up modems later used phone lines for access (tunneling data over circuits), the core internet protocol (TCP/IP) is packet-based, not circuit-based.
In summary, circuit switching excels at constant-bit-rate voice but is rigid and wasteful for data. Packet switching’s flexibility made the modern internet possible.
References:
- https://en.wikipedia.org/wiki/Public_switched_telephone_network
- https://en.wikipedia.org/wiki/Packet_switching
- https://en.wikipedia.org/wiki/ARPANET
- https://www.nextiva.com/blog/what-is-pstn.html
- https://ethw.org/Packet_Switching
- https://en.wikipedia.org/wiki/Dial-up_Internet_access