Fault Tolerance

The expectation is that the Internet is always available to the millions of users who rely on it. This requires a network architecture that is built to be fault tolerant. A fault tolerant network is one that limits the impact of a failure, so that the fewest number of devices are affected by it. It is also built in a way that allows quick recovery when such a failure occurs. These networks depend on multiple paths between the source and destination of a message. If one path fails, the messages can be instantly sent over a different link. Having multiple paths to a destination is known as redundancy.

Circuit-Switched Connection-Oriented Networks

To understand the need for redundancy, we can look at how early telephone systems worked. When a person made a call using a traditional telephone set, the call first went through a setup process. This process identified the telephone switching locations between the person making the call (the source) and the phone set receiving the call (the destination). A temporary path, or circuit, was created for the duration of the telephone call. If any link or device in the circuit failed, the call was dropped. To reconnect, a new call had to be made, with a new circuit. This connection process is referred to as a circuit-switched process and is illustrated in the figure.

Many circuit-switched networks give priority to existing circuit connections at the expense of new circuit requests. After a circuit is established, even if no communication is occurring between the persons on either end of the call, the circuit remains connected and resources used until one of the parties disconnects the call. Because there are only so many circuits that can be created, it is possible to get a message that all circuits are busy and a call cannot be placed. The cost to create many alternate paths with enough capacity to support a large number of simultaneous circuits, and the technologies necessary to dynamically recreate dropped circuits in the event of a failure, is why circuit switched technology was not optimal for the Internet.