Memory Management Method for Power Failure Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for determining invalid data in memory storage apparatuses due to abnormal power failures often incorrectly identify valid data as invalid, leading to errors, as they do not accurately distinguish between errors caused by power failures and those from other factors like data retention or interference.
Innovation Solution
A memory management method that identifies a first physical programming unit in a fully-erased status and compares its status data with that of previously programmed units, computing differences to determine if the second physical programming unit is in a program failure status by applying predetermined read voltages and using error checking and correcting codes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all pages are scanned and error bits are counted to determine invalid data, then power failure detection is implemented, but valid data with high error bits is incorrectly identified as invalid data
Solution Approach 1:
The controller separates memory pages into candidate pages and reference pages based on their programming order. The reference page, being programmed after the candidate pages, serves as a control sample with expected normal error characteristics. This segmentation enables accurate comparison to identify power failure cases without misclassifying valid data
Solution Approach 2:
The controller counts error bits in both candidate pages and reference pages, then compares these counts to determine validity. The reference page error count provides feedback on normal error levels, allowing the system to identify when candidate pages have abnormally high errors indicating power failure, versus when high errors are within normal ranges
2Object-affected harmful factors
If conventional error detection methods are used, then power failure errors are detected, but data integrity is compromised due to false identification of valid data as invalid
Solution Approach 1:
The controller first identifies and classifies pages into candidate and reference groups based on programming order before performing error bit counting and comparison. This preliminary classification action ensures that the subsequent error detection process has the proper structural framework, allowing accurate distinction between power failure errors and normal degradation without compromising data integrity
Data Source
AI summary
The present disclosure provides a memory management method for a rewritable non-volatile memory module. The rewritable non-volatile memory module includes physical programming units, each of which includes multiple bits. The memory management method includes: identifying a first physical programming unit by applying a predetermined read voltage, where the first physical programming unit is identified as in a fully-erased status; identifying a second and a third physical programming units which are programmed before the first physical programming unit; acquiring status data of the second and the third physical programming unit; computing a difference of the status data between the second and the third physical programming unit; if the difference is larger than a threshold, identifying the second physical programming unit as in a program failure status.