Fault-tolerant system and method for adjusting error correcting codes adaptively

Abstract

The invention discloses a fault-tolerant system and method for adjusting error correcting codes adaptively. The fault-tolerant system comprises an environmental information sensing module, a fault-tolerant scheme selection module, an error correcting module and an object processing module. The environmental information sensing module is connected with the fault-tolerant scheme selection module and is used for obtaining information of factors which influence the reliability of data stored in the object processing module. The fault-tolerant scheme selection module is connected with the environmental information sensing module and the error correcting module and selects a mould generation and output fault-tolerant scheme according to the received reliability influence factors from the environmental information sensing module and the feedback object processing module health degree from the error correcting module. The error correcting module receives the fault-tolerant scheme transmitted from the fault-tolerant scheme selection module and is used for defecting and correcting the errors in the data transmitted by the object processing module. The object processing module is connected with the error correcting module and is used for initiating data read-write control operation, sending original data and tasks to the error correcting module and storing reliable data output by the error correcting module.

Description

Technical field [0001] The present invention relates to the field of intelligent fault tolerance of storage devices, in particular to the field of adaptive fault tolerance control using error correction codes. Background technique [0002] Current storage devices will more or less fail during use, and cannot fully guarantee the reliability of stored data. Data read and write errors may occur, which affects the reliable operation of the system. At present, the chip-level fault-tolerant solutions mainly adopt TMR three-mode redundancy technology and error correction control coding technology. Among them, the TMR three-mode redundancy technology needs to store the same data in three devices, which causes a great waste of resources and power consumption, and is generally only used for key control locations. The use of error correction control codes only needs to add a few redundant bits to achieve the purpose of data error detection and correction, and has a higher use efficiency. ...

AI Technical Summary

Problems solved by technology

Various error-correcting codes have their own advantages and disadvantages. The traditional method of use is artificially pre-set. After setting, it cannot be changed, and the use of error-correcting codes cannot be adjusted according to the actual environment.

The complexity of several error correction codes varies greatly. If only one error correction code is used to ensure reliability, it will cause a great waste of resources in most low-risk areas, which is not conducive to the improvement of chip usage efficiency. Contrary to the current trend of faster and faster data transfer rates

Images