Well, for example, selling tickets to the train. Every day, many people do the same operation. Main characteristics: fixed operations (the ticket looks almost the same for everyone) and a large flow of input data. It is clear that if you use the traditional (as with sites, for example) DBMS approach without doping, this will be extremely inefficient, since Most of the functionality will generally stand idle, but it also does not interfere where it is asked. The OLTP system still has, or may not, have a traditional DBMS, but already focused on speed. How can speed be increased — predefined sizes of requests, an enumerable set of fields (zone, direction, final, etc.), a relatively simple structure of the storage itself (so that it does not have to perform three-story requests), simple integrity control mechanisms (transactionality). From here and online - on the fly, transaction - all actions should be brought to the end, processing - processing :). In general, OLTP is a concept that can define, help, and suggest how best to implement fast processing of all large data streams. Something like this