Storage Controller Gray Code Bias Detection for Abnormal Memory Cells
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional memory management methods for rewritable non-volatile memory modules inaccurately identify damaged word-lines, leading to excessive reduction in available storage space as all word-lines in a block are deemed unusable even if only some are damaged, resulting in wasted space and decoding errors.
Innovation Solution
A memory management method and storage controller that utilize Gray code bias values to identify abnormal memory cells by comparing raw and decoded data, calculating Gray code absolute bias values, and recording abnormal cells, thereby precisely determining damaged cells and improving storage efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional methods determine a physical block as bad when any word-line is damaged, then decoding errors are reduced, but storage space utilization deteriorates due to excessive reduction in available space
Solution Approach 1:
The patent segments the physical block into individual word-lines for independent evaluation. Instead of treating the entire block as bad when any word-line is damaged, the system identifies and isolates only the specific damaged word-lines, allowing healthy word-lines to remain usable. This segmentation resolves the contradiction by maintaining reliability through error detection while preserving storage space by not unnecessarily marking entire blocks as bad.
Solution Approach 2:
The patent applies local quality assessment by evaluating each word-line's health status independently rather than applying a uniform bad status to the entire block. The system identifies specific local defects (damaged word-lines) while preserving the quality and usability of unaffected areas (healthy word-lines), thus resolving the contradiction between ensuring reliability and maximizing storage space utilization.
2Reliability
If all word-lines in a block are deemed unusable when some are damaged, then data integrity is improved, but storage efficiency deteriorates due to wasted space
Solution Approach 1:
The patent divides the block into separable word-lines and evaluates each independently. This segmentation allows the system to maintain data integrity by identifying and isolating damaged word-lines while preserving healthy ones for storage operations, thereby resolving the contradiction between ensuring data integrity and maintaining storage efficiency.
Solution Approach 2:
The patent selectively discards only the damaged word-lines while recovering and preserving the healthy word-lines for continued use. This selective discarding and recovery approach resolves the contradiction by maintaining data integrity through removal of defective elements while preserving storage efficiency by keeping functional elements available for data storage.
3Reliability
If conventional methods mark entire physical blocks as bad, then abnormal memory cells are isolated, but storage capacity deteriorates due to excessive space reduction
Solution Approach 1:
The patent segments the block into individual word-lines and isolates only the abnormal ones rather than marking the entire block as bad. This selective isolation approach resolves the contradiction by maintaining reliability through abnormal cell identification while preserving storage capacity by keeping healthy word-lines available for use.
Solution Approach 2:
The patent applies local quality assessment to identify and isolate specific abnormal word-lines while preserving the quality and usability of healthy word-lines. This local isolation strategy resolves the contradiction between ensuring reliability through abnormal cell separation and maintaining storage capacity by not unnecessarily reducing available space.
Data Source
AI summary
A memory management method and a storage controller using the same are provided. The method includes reading a target word-line to identify a plurality of raw Gray code indexes corresponding to a plurality of memory cells of the target word-line; performing a decoding operation on raw data of the target word-line to identify a plurality of decoded Gray code indexes corresponding to the memory cells; calculating a plurality of Gray code absolute bias values corresponding to the memory cells according to the raw Gray code indexes and the decoded Gray code indexes; and identifying one or more abnormal memory cells among the memory cells according to the Gray code absolute bias values; and recording the one or more abnormal memory cells into an abnormal memory cell table, wherein a Gray code absolute bias value of each of the one or more abnormal memory cells is greater than a bias threshold.