Abstract: A back-up control system of a redundancy configuration suited to be used for a time division switching network. The control system includes connecting paths extending from each of the load supply devices, the paths being classified into an active connecting path and a corresponding standby connecting path coupled to different ones of the load accommodating devices which are provided in a number equal to or larger than three. The total load accommodating capacity of the load accommodating devices is provided with a margin or excess capacity which is smaller than the capacity required for accommodating all full loads supplied from the load supply devices but is not smaller than the load accommodating capacity of any one of the load accommodating devices. At least one of the load accommodating devices is connected to the active connecting path and the spare connecting path which corresponds to another one than the just mentioned active connecting path.

Abstract: A multiprocessor information processing system has an additional function to detect a fault occurring in a processor. A counter is provided for each of a plurality of processors for holding an associated fault supervising code. The code stored in the associated counter is periodically updated by the associated processor while the update status of the code is supervised on a cycle longer than the cycle of the updating period. If a fault occurs in one of the processors, the fault supervising code corresponding to that processor will not be updated. Thus, the faulty processor can be detected by periodically supervising the update status of the fault supervising code. The supervising operation can be carried out by software or hardware.