A fault self-healing system based on system reconfiguration and its implementation method

Abstract

The invention proposes a system reconfiguration based fault self-recovery method. An applied computer system consists of a plurality of functional modules and a backup module in hardware and consists of a fault management module, a configuration management module, a system blueprint module and other modules in software. The method sequentially comprises the steps of system starting, configuration loading, fault detection, fault recording, fault filtering, policy selection, configuration updating and system shutdown. A system blueprint module is a key design of system fault self-recovery and comprises configuration items run by a plurality of systems; each configuration item describes various configurations of software, hardware, a network and the like of the computer system; the computer system can run according to the descriptions of the configuration items; and the configuration items further comprise descriptions of all fault handling policies. The backup module is a necessary condition for system self-recovery and can work instead of a module when the module has an unrecoverable fault in the system. The method has the advantages of simple self-recovery policy, controllable self-recovery process and definite self-recovery result.

Description

technical field [0001] The invention belongs to the field of computers and provides a fault self-healing method. Background technique [0002] In some complex computer systems that are frequently used but difficult to maintain, such as civil aircraft onboard electronic systems and satellite onboard electronic systems, designers hope to improve system availability and reduce system maintenance cycles and costs through system self-healing design the goal of. [0003] The self-healing technology currently used mainly achieves the purpose of self-healing of the system function through the software and hardware recovery of the fault location. Some common technical measures are: [0004] (1) Design multiple channels at the location where a fault may occur, and select a non-faulty channel to bypass the faulty channel after a fault occurs, so as to realize the self-healing of the system; [0005] (2) When the software code and FPGA logic stored in the permanent memory are destroye...

AI Technical Summary

Problems solved by technology

[0006] The above self-healing technology is mainly designed for the recovery of local functions of the system. Although it has a good effect, many hardware failures still cannot realize the self-healing of functions.

Images