A key quality of any kind of system is its ability to deliver its respective service correctly. Often the unavailability of commercial systems may lead to lost revenue, which are minor compared to what may be at stake when critical infrastructures fail. A failure to deliver critical services, such as clean water or electricity may have dire consequences that endanger human lives and may even halt or break other infrastructures. The services provided by critical infrastructures need to be supplied continuously even when faced with re-configurations, outside disturbances and systemic changes. A system is called resilient if it fulfils this property. From the many critical infrastructures that exist, power systems may be the most important ones, because they are supplying the required electricity for other critical infrastructures. At the same time, a power system itself may be exposed to several disturbances from internal sources, e.g., fluctuations in the energy demand, and external sources, e.g., heavy storms. Especially, fast dynamic effects caused by these disturbances may lead to deviations of grid frequency, short-circuits, or, in severe cases, a total power system failure. As future scenarios will include more distributed renewable sources and less centralized generation from fossil fuels, ICTbased communication and coordination will play an increasing role. This paper examines the notion of resilience, how it has been traditionally ensured for the power system, and novel approaches to maintain the frequency, protect people and devices against short circuits and recover from a blackout. A special focus is on communication and the role that distributed renewable generation plays for these processes.