Predicting At-Risk Blocks in Non-Volatile Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing non-volatile storage systems cannot effectively detect and address memory blocks that deteriorate over time, leading to gradual loss of data or inability to read contents, even if the programming process is successful, without being accessed or accessed only for reading.
Innovation Solution
A method is introduced to predict at-risk blocks in non-volatile storage systems by analyzing performance data, reprogramming data from these blocks to safer locations, and marking them as bad to prevent further use, utilizing techniques such as multiple testing phases and monitoring programming loop counts to identify potential failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If stress testing and programming verification are performed during manufacturing, then manufacturing defects are detected, but gradual deterioration after successful programming cannot be detected
Solution Approach 1:
The patent performs preliminary actions by conducting multiple read operations and analyzing programming loop counts immediately after programming to predict future deterioration. This early detection mechanism identifies blocks that will deteriorate before actual data loss occurs, enabling proactive data relocation and preventing information loss.
Solution Approach 2:
The patent implements feedback mechanisms by continuously monitoring programming loop counts and read operation results. This feedback loop provides information about block health status, allowing the system to identify deteriorating blocks and trigger data relocation before complete failure occurs, thus preventing data loss.
2Reliability
If multiple testing phases and programming loop count monitoring are implemented, then at-risk blocks are identified early, but system complexity increases
Solution Approach 1:
The patent applies self-service by utilizing existing programming and read operations to gather diagnostic information. The programming loop count, which is already generated during normal programming, is repurposed as a health indicator. This eliminates the need for separate complex testing hardware while still achieving reliable failure prediction.
Solution Approach 2:
The patent makes existing components multi-functional by using the programming loop count mechanism for both its original purpose (controlling programming completion) and as a diagnostic tool for predicting block deterioration. This universal use of existing data structures and operations reduces overall system complexity while improving reliability.
3Reliability
If data is relocated from at-risk blocks to safe blocks, then data integrity is maintained, but storage capacity is reduced
Solution Approach 1:
The patent discards deteriorating blocks from active use and recovers their data by relocating it to healthy blocks. This process maintains data integrity while managing storage capacity efficiently. The discarded blocks can potentially be recovered through refurbishment processes, and the system continuously monitors and reallocates storage resources to maximize usable capacity.
Data Source
Figure 1~7
Figure 3
Figure 4
AI summary
The non-volatile storage system predicts which blocks (or other units of storage) will become bad based on performance data. User data in those blocks predicted to become bad can be re-programmed to other blocks, and the blocks predicted to become bad can be removed from further use.