Events provide a fundamental abstraction for representing time-evolving information. Complex event recognition focuses on tracking and analysing streams of events, in order to detect patterns of special significance. The event streams may originate from various types of sensor, such as cameras and GPS sensors. Furthermore, the stream velocity and volume pose significant challenges to event processing systems. We propose dRTEC, an event recognition system that employs the Event Calculus formalism and operates in multiple processing threads. We evaluate dRTEC using two real-world applications and show that it is capable of real-time and scalable event recognition.