Flash Memory Re-Encoding for Long-Term Data Retention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As lithographic geometry for NAND flash chips decreases, the number of electrons used in storage decreases, making it difficult to ensure accurate storage of information, and there is a need for improved data retention characteristics in flash memory systems, especially for long-term data preservation without frequent access or power.
Innovation Solution
A method and system that transition flash memory to a long-term data retention state by re-storing encoded data at a lower code rate, involving reading, decoding, and re-encoding data, while also performing error correction and threshold voltage selection to optimize data storage and retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multi-level cells are used to increase storage density, then storage capacity is improved, but data retention reliability deteriorates
Solution Approach 1:
The patent applies parameter changes by transitioning the code rate parameter from a first code rate to a second code rate that is lower than the first code rate. This change in the code rate parameter increases the redundancy of error correction codes, thereby improving data retention reliability while maintaining compatibility with multi-level cell storage structures
Solution Approach 2:
The patent implements beforehand cushioning by re-storing encoded data with enhanced error correction capabilities before potential data loss occurs. The process involves reading existing data, decoding it, re-encoding with a lower code rate (higher redundancy), and writing it back, thereby preparing the data with additional protection against future retention failures
2Quantity of substance
If smaller memory cells are used to reduce lithographic geometry, then storage density is improved, but manufacturing precision deteriorates
Solution Approach 1:
The patent introduces an intermediary mechanism in the form of error correction codes that mediate between the physical limitations of small memory cells and the requirement for accurate data storage. The ECC acts as a buffer that compensates for the reduced precision in electron storage by providing algorithmic correction capabilities
Solution Approach 2:
The patent changes the code rate parameter to a lower value, which increases the proportion of redundant check bits in the stored data. This parameter change effectively compensates for the manufacturing precision limitations of smaller cells by adding more error detection and correction capability
3Productivity
If higher code rate is used to maximize storage capacity, then storage efficiency is improved, but error correction capability deteriorates
Solution Approach 1:
The patent applies dynamics by making the code rate adjustable rather than fixed. The system can dynamically transition between a first code rate optimized for storage efficiency and a second, lower code rate optimized for data retention and error correction, allowing the storage system to adapt to different operational requirements
4Reliability
If stronger error-correction code is applied to compensate for reduced signal-to-noise ratio, then data reliability is improved, but device complexity increases
Solution Approach 1:
The patent applies partial action by selectively re-storing data with enhanced error correction only when needed for long-term retention, rather than always using the highest level of error correction. This allows the system to balance reliability requirements with processing complexity by applying stronger ECC only in retention-critical scenarios
Data Source
AI summary
A system and method for long term data retention in a flash memory. In some embodiments, the method includes transitioning the flash memory to a long term data retention state by re-storing first encoded data, the first encoded data being initially stored in the flash memory at a first code rate. The re-storing may include determining a second code rate, lower than the first code rate; reading the first encoded data from the flash memory; decoding the first encoded data at the first code rate to obtain first decoded data; encoding the first decoded data at the second code rate to form second encoded data; and storing the second encoded data in the flash memory.