Events with significance to the system are identified within an event-driven program. An event could be some user action, a transmission of sensor data or a message from some other program or system, among an almost infinite number of other possibilities. The ECA rule specifies how events drive the desired program responses. When an event with significance for the system occurs, the conditions are checked for or evaluated; if the conditions exist or meet pre-established criteria, the appropriate action is executed.
ECA rules originated in active databases and have since been used in areas including personalization, big data management and business process automation. The model is being explored for M2M (machine-to-machine) networking, Internet of Things (IoT), cognitive computing and the semantic web.
See an introduction to event-driven fundamentals from Linux Academy's Lambda Deep Dive course: