NAND Memory Dual Refreshing Method for Lifespan Extension
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In NAND flash memory systems, the existing refreshing processes lead to accumulated damage and reduced lifespan due to the necessity of erasing data, which is exacerbated by the increasing number of error bits over time, causing data loss when error correction fails.
Innovation Solution
Implementing a dual refreshing method where normal refreshing moves data to another block and overwrite refreshing reprograms error cells without erasing, thereby reducing damage and power consumption, and extending the lifespan of the NAND memory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If normal refreshing is executed to move data to another block, then data integrity is maintained, but erasing operations accumulate damage and reduce memory lifespan
Solution Approach 1:
The refreshing process is segmented into two distinct types: normal refreshing for blocks with low error bits and overwrite refreshing for blocks with high error bits. This segmentation allows each type to be optimized independently, with overwrite refreshing avoiding erasing operations to preserve memory lifespan while normal refreshing maintains data integrity when necessary.
Solution Approach 2:
The system dynamically changes the refreshing method based on the error bit count parameter. When error bits exceed a threshold, the system transitions from normal refreshing (which performs erasing) to overwrite refreshing (which skips erasing), thereby adapting the process to minimize damage accumulation while maintaining reliability.
2Reliability
If normal refreshing is executed with erasing operations, then data is refreshed reliably, but power consumption increases and operation speed decreases
Solution Approach 1:
The overwrite refreshing process performs only the necessary reprogramming operation without the full erasing sequence. By applying partial action (skipping the erasing step when error bits are high), the system reduces power consumption and operation time while still achieving the essential goal of data refresh through direct reprogramming.
3Loss of energy
If overwrite refreshing is executed for blocks with high error bits, then damage and power consumption are reduced, but data loss risk increases if error correction fails
Solution Approach 1:
The system uses the error bit count as a dynamic parameter to determine the refreshing method. By monitoring this parameter and comparing it against thresholds, the system intelligently selects between normal refreshing and overwrite refreshing, balancing the trade-off between power consumption and data loss risk based on the actual state of the memory block.
Data Source
AI summary
According to one embodiment, a controller executes first refreshing in a case where a first value of a first block is larger than a first threshold and less than a second threshold. The first refreshing includes reprogramming a plurality of second memory cells among a plurality of first memory cells included in the first block.


