Survey on Multiprocessor Fault-Tolerant Scheduling in Safety Critical System

As tasks in the fields of aerospace and industrial automation become progressively more complex and intelligent, higher demands are placed on the computing power of computers. In order to integrate more functions or services in hardware to meet market demands, computer systems are beginning to use multiprocessors or multi-core processors. For real-time systems containing hard real-time tasks, the system must be not only timeliness but also fault tolerance. Timeliness means that tasks are completed before the deadline and fault tolerance means that the correct output is obtained within the deadline even in the event of a fault. Current research on fault-tolerant computer technology is gradually being extended to fault-tolerant scheduling of multiprocessors or multi-core processors. The implementation of fault-tolerant systems relies primarily on redundancy. One classification of fault-tolerant methods is time, spatial and information redundancy, another classification is hardware redundancy and software redundancy. This paper summarizes fault-tolerance techniques and provides an overview of fault-tolerant scheduling algorithms for multiprocessors or multi-core processors.