A method for monitoring and managing redundancy of a fault-tolerant system for improving availability

Abstract

The invention discloses a redundancy management method for improving the availability of a monitoring pair based fault tolerant system. The method comprises the steps of intra-node voting and monitoring and node-level voting and monitoring. During the intra-node voting and monitoring, a controller A and a controller B are consistent by comparison and data of the controller A are selected to enter the node-level voting and monitoring. During the node-level voting and monitoring, a mid-value is selected as a reference value of monitoring from values of three signals of three nodes, and if a difference between other two non mid-value signals and a mid-value signal is within a threshold range, then the non mid-value signals are normal, otherwise, the non mid-value signals are faulted. According to the redundancy management method, the redundancy management of a monitoring pair based three-node or two-node fault tolerant computer is realized; the credibility of system data of a fault tolerant computer system can be ensured; and compared with fault silence based monitoring pair fault tolerance, the redundancy degradation of the system is slowed down, and a backup controller can be reconfigured to be a main controller when a fault occurs once in a node through a software and hardware combination mode, so that the system is ensured not to be degraded.

Description

technical field [0001] The technology belongs to the technology and application of a highly reliable airborne fault-tolerant computer. Background technique [0002] Long-endurance unmanned aerial vehicles have the characteristics of long flight time and high requirements for reliability and availability. Usually to improve work reliability and availability, multi-node fault-tolerant computers with fault-tolerant capabilities are usually used to achieve reliable execution of core functions, such as two-node fault-tolerant computers with one-time failure safety, three-node fault-tolerant computers with one-time failure / working ability, and two-time failure Four-node fault-tolerant computers with / working capacity, etc. Fault-tolerant is to use redundant components or components to shield the impact of faults on the system, but it needs to increase the resource overhead of software and hardware in the system. Therefore, fault tolerance can be defined as: If a system can rely o...

AI Technical Summary

Problems solved by technology

Usually to improve work reliability and availability, multi-node fault-tolerant computers with fault-tolerant capabilities are usually used to achieve reliable execution of core functions, such as two-node fault-tolerant computers with one-time failure safety, three-node fault-tolerant computers with one-time failure / working ability, and two-time failure Four-node fault-tolerant computers with / working capacity, etc. Fault-tolerant is to use redundant components or components to shield the impact of faults on the system, but it needs to increase the resource overhead of software and hardware in the system

Images