The design of object-oriented real-time distributed systems is especially hard due also to the fact that some issues (distribution) are dealt with too early in the system life-cycle, others (strategy) are hard-coded as programming-in-the-small issues, others (time) are hidden inside the operating system kernel. We present a reference architecture for monitoring and control systems that addresses the first two of these issues. The architecture lifts distribution, activation and timing up to an explicit, programming-in-the-large level.