NAND Flash Block Classification for Error-Aware Life Extension
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The durability and reliability of NAND flash memory are compromised due to the impact of erase/write operations, particularly with the thinning of the oxide layer, leading to misclassification of physical units as faulty, thereby reducing the service life of memory storage devices.
Innovation Solution
A memory control method that classifies physical units into types based on execution counts and operational conditions, temporarily avoiding misclassification during environmental disturbances, and performing re-evaluation in idle states to extend the service life.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If physical units are marked as faulty after errors occur during operations, then data reliability is improved, but service life is reduced due to misclassification during environmental disturbances
Solution Approach 1:
The system performs preliminary actions by checking execution counts and operation types before marking physical units as faulty. It evaluates whether errors occurred during critical operations (erase/write) versus read operations, and whether the unit has reached its endurance threshold, thereby preventing premature misclassification and extending service life while maintaining data reliability
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring execution counts, operation types, and error occurrences. This feedback loop allows the system to dynamically adjust marking decisions based on actual usage patterns and environmental conditions, reducing false positives during temporary environmental disturbances while ensuring reliable data protection
2Reliability
If execution count thresholds are set low to ensure data reliability, then reliability is improved, but productivity is reduced due to premature marking of usable units
Solution Approach 1:
The system changes parameters by dynamically evaluating execution counts against operation-specific thresholds rather than using a single fixed threshold. It distinguishes between read count, write count, and erase count, applying different threshold criteria for each operation type, thereby maximizing storage availability while ensuring data reliability through appropriate parameter selection
3Reliability
If all physical units are marked as faulty after any error, then data reliability is improved, but device complexity increases due to unnecessary classification management
Solution Approach 1:
The system applies local quality by treating different physical units differently based on their specific error history and usage patterns. Rather than uniformly marking all units as faulty after any error, it applies marking decisions locally to specific units that meet defined criteria (execution count thresholds, operation type, error frequency), thereby maintaining simplicity while ensuring reliability
Data Source
AI summary
A memory control method and a memory storage device may extend service life and ensure reliability of data. The memory control method is used for a rewritable non-volatile memory module including multiple physical units, and includes: when an error occurs in a target operation executed on a first physical unit, obtaining an execution count of the first physical unit; if the execution count and operation type indicate that the first physical unit satisfies the first preset condition, determining whether the first physical unit is the third type; if the first physical unit is not the third type and the execution count indicates that the first physical unit satisfies the second preset condition, determining whether the first physical unit is the second type; and if the first physical unit is not the second type, marking the first physical unit as the second type.