The Design And Implementation Of Fault-tolerant Based On Minicore System

The computer system in space environment may occur single event upset by the influence of environment, at the same time also will encounter some hardware and software errors, such as component failure, software bugs, so some measures must be taken to improve the reliability of the system. Fault tolerance is an effective means to improve the reliability of computer systems, as an important aspect of fault-tolerant technology, software fault tolerance to improve the system provides a flexible and effective way to reliability.Contrary to limited storage space for embedded real-time systems, this paper was aimed to provide a safe and reliable mechanism for the fault-tolerant computer systems in the software level. The Target is to provide a fault tolerant protection mechanisms at the system level which will guarantee the system data’s accuracy and improve the success rate of service execution system and to provide a configurable fault tolerance mechanisms in user level which users can selecting fault-tolerant strategies depending on the application characteristics at the same time. For the above objective, this paper mainly completed the work in three areas as follows:First introduced real-time fault-tolerant system’s current research status, and describes the use of software and hardware fault tolerance in the operating system. Then, proposed a two fault-tolerant design at the system level and user-level.Among them, the system-level fault-tolerant design consists of an information redundant strategy based on memory and a triple modular redundancy strategy based on system services, the user-level fault-tolerant design consists of a master/slave version fault-tolerant scheduling strategy and a N-versions fault-tolerant scheduling strategy. The third is proposed a modification fault-tolerant scheduling algorithm.Finally, in order to verify the effectiveness of the proposed software fault tolerance techniques, the experimental verification of the program carried out in an embedded operating system named MiniCore which based on a service body/ execution flow model (SEFM).Experimental results show that the introduction of fault-tolerant mechanism increases the code size of Minicore kernel by 33% and declines the system’s time performance slightly, in the simulation under the influence of single event upset (SEU), ensured the correctness of the memory data and the success rate of task execution, and improved the reliability of computer system.