An approach to the support of the fault-tolerance of the double redundant computer control systems

Abstract

Faults in computer control systems cause great economic losses and endanger human beings. In order to ensure control and monitoring tasks, the availability and safety of computer control systems in areas such as military, manufacturing plants, traffic, public health service, environment protection, banking and assurances have to be improved. The fault-tolerance of the computer control system is achieved by using the hardware and software redundancy and redundancy of processor time. In this paper, the structure of real time double redundant computer control system is updated. The highly efficient methods and device are designed for supporting of the fault-tolerance of the double redundant computer control system by using the hardware redundancy. The fault-tolerant system software enables identical application programs to work parallel on two reserved computers. The computational process executed in a computer control system is periodically interrupted at checkpoints by execution of the fault-tolerant procedure, which recognizes and eliminates the faulty computers. The recovery of computer control system is done automatically without interruption of the control services.