SSD Block Refresh for Read Retry Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional solid state storage devices experience increased read retry possibilities due to shifted threshold voltage distribution curves over time, leading to decreased reading speed and potential data loss, especially as program erase count increases.
Innovation Solution
Implement a control method that periodically checks memory blocks for data storage time and program erase count, predicts read retry possibilities using block and environment information, and tags or refreshes data in blocks with high read retry probabilities to maintain accurate data retrieval and prevent read failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional solid state storage devices are used without proactive management, then device complexity is reduced, but read retry possibility increases and reading speed decreases over time
Solution Approach 1:
The patent applies preliminary action by proactively monitoring memory block characteristics (program erase count, data storage time) and predicting read retry possibilities before actual read failures occur. This allows the system to refresh or relocate data from high-risk blocks in advance, preventing read failures and maintaining high read success rates without adding complex real-time intervention mechanisms.
2Reliability
If read retry operations are performed frequently, then data retrieval accuracy is maintained, but reading speed decreases
Solution Approach 1:
The patent implements feedback by continuously monitoring memory block characteristics such as program erase count and data storage time, and using this information to predict read retry possibilities. The system adjusts its behavior based on this feedback by proactively refreshing or relocating data from blocks with high predicted read retry probabilities, thereby maintaining data retrieval accuracy while avoiding the performance penalty of frequent reactive read retry operations.
3Productivity
If memory blocks are monitored and managed proactively, then read retry possibility is reduced, but control complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the memory into manageable blocks and independently monitoring key characteristics of each block (program erase count, data storage time). This segmented approach allows the system to apply proactive management selectively to high-risk blocks without unnecessarily complicating the control of the entire memory system, thus improving reading speed while keeping control complexity manageable.
4Reliability
If data in high-risk blocks is refreshed or relocated, then data integrity is maintained, but additional write operations increase
Solution Approach 1:
The patent applies parameter changes by using predictive analytics based on monitored parameters (program erase count, data storage time, block characteristics) to identify high-risk blocks. Instead of universally refreshing all blocks, the system selectively refreshes or relocates data only from blocks with high predicted read retry probabilities, thereby maintaining data integrity while minimizing the additional write operations and associated energy consumption.
Data Source
AI summary
A control method of a solid state storage device includes the following steps. Firstly, a block of a memory cell array is checked. Then, a judging step is performed to judge whether a data storage time period of the block exceeds a threshold period. If the data storage time period of the block exceeds the threshold period, the block is tagged or a data of the block is refreshed.