NAND Flash Read Retry Using Precomputed Re-Read Voltages
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Flash memory devices experience read errors due to voltage drift, leading to instability and data loss, especially in NAND flash devices, which are not random access and require serial access commands, making existing read retry mechanisms inefficient.
Innovation Solution
Implement a read retry mechanism in the flash controller that uses a set of re-read voltages from a voltage difference table to correct read errors, including a processing unit that adjusts re-read voltages based on error correction units to ensure accurate data retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional read retry mechanisms are used in flash memory devices, then read errors can be corrected, but device-side delay and power consumption increase
Solution Approach 1:
The patent pre-calculates and stores voltage difference tables during manufacturing or initialization, containing optimal voltage adjustments for different error patterns. During read retry operations, the controller directly references these pre-computed tables rather than performing complex real-time calculations, significantly reducing the time required for error correction while maintaining high reliability
Solution Approach 2:
The system performs self-diagnosis by detecting read errors and automatically selecting appropriate correction voltages from the pre-stored tables based on error patterns. The flash controller autonomously manages the retry process with minimal host intervention, reducing both delay and power consumption compared to traditional mechanisms that require more extensive host-controller communication
2Reliability
If traditional read retry mechanisms are used in flash memory devices, then read errors can be corrected, but power consumption increases
Solution Approach 1:
By pre-computing and storing voltage difference tables during initialization, the patent eliminates the need for power-intensive real-time calculations during read retry operations. The controller only needs to perform simple table lookups and voltage applications, dramatically reducing power consumption while maintaining effective error correction
Solution Approach 2:
The system applies only the necessary voltage adjustments from the pre-stored tables based on the detected error patterns, rather than attempting all possible corrections. This selective approach corrects errors efficiently with minimal power expenditure, avoiding unnecessary voltage cycling that would increase power consumption
3Reliability
If re-read voltages are adjusted during read retry, then read error correction improves, but read performance decreases due to increased processing steps
Solution Approach 1:
The voltage difference tables are pre-calculated and stored during manufacturing or initialization, containing optimal voltage adjustments for various error scenarios. During read retry, the controller performs only simple table lookups and voltage applications, maintaining high read performance while achieving accurate error correction through the pre-optimized voltage sequences
Data Source
AI summary
The invention introduces a method for read retry, performed by a processing unit, includes: providing a first set of re-read voltages in an iteration of a read retry mechanism; obtaining a second set of re-read voltages with a maximum distance from the first set of re-read voltages; driving a flash interface (I/F) to use the second set of re-read voltages to read a designated page and obtain second data from a flash module; and in response that the second data is correct, or the second data is successfully corrected by an error correction unit, exiting the read retry mechanism and replying with the second data or corrected second data to a host side or storing the second data or the corrected second data at a designated address of a random access memory (RAM).