All currently existing telecommunication networks can use two types of communication to provide the required functions - switching channels and (or) packets. What is it and how do they differ from each other?
Let's start with how circuit-switched networks work. They appeared earlier than their analogs with dial-up packages, therefore, which is not surprising, they are easier to implement. A striking example of a network that uses circuit switching is the telephone line. Obviously, in order for two subscribers to start communication, it is necessary to establish a connection between them. The initiating subscriber dials a number, which, in fact, is a command to the telephone exchange equipment (PBX) located between them, to properly connect two lines - from the initiator and from the responder (we take an example when subscribers are served by one station). Previously, mechanical probes on the simplest position sensors were used for this , then, with the advent of digital solutions, the implementation changed, although the principle remained the same. Switching channels provides subscribers with an independent line, which remains assigned to them until the end of the communication session. The advantages are obvious: high reliability, no need to transmit control packets. However, this connection method with an increase in the number of subscribers becomes too wasteful, since the number of channels is physically limited. Even an attempt to solve this problem by applying seals is only a temporary measure, defined by an interim solution. In addition, channel switching has one significant drawback - the communication line is busy all the time, even if there is no exchange of information between subscribers. For example, during a telephone conversation, you can hang up next to the device and go about your business - the channel will remain reserved for them until you receive a signal to disconnect.
That is why subsequently the switching of channels was replaced by a method of packet switching. The principle of its operation involves encoding and splitting the transmitted data stream into a number of separate packets, which are transmitted to the recipient through a common communication line and are combined there into the original stream. To understand the differences between these two methods, you can use the analogy with the transport line: when switching channels, the line is represented by a railway track, and the data stream is a train of many cars. It is quite clear that delays are extremely rare along the route, and reliability is one of the highest. At the same time, several trains cannot move along this track at the same time. But packet-switched lines are a high-speed highway with multi-lane traffic. The transported cargo (transmitted packages) is divided into several machines which, tacking in the stream of other modes of transport, reach their destination, where the initial structure is assembled. In this example, a road is a communication channel, and cars are data packets. They coexist quietly on the same road, almost without interfering with each other's movements. Exceptions are traffic jams, traffic lights and emergency situations (these are delays). Even if any machine did not arrive at the recipient, a copy of it on request can be sent again. The total amount of information transmitted per unit time during packet switching is significantly higher than in the case of channels.
In general, switching is switching something, changing states. In network technology, it forms a route for passing data. A feature is the way it is organized. Switching should not be confused with routing, whose task is to find the optimal route.