Storage Device Retention Error Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Flash memory devices experience read performance degradation due to retention characteristics, leading to excessive error bit recovery operations that can persist until data relocation occurs, causing continuous severe read performance issues.
Innovation Solution
A method for operating a storage device that selects a recovery scheme to recover performance degradation by detecting error levels in memory cells and relocating data from blocks with a threshold amount of errors to maintain data integrity and improve read performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data relocation is performed to recover performance degradation, then read performance is improved, but device complexity and operation time increase
Solution Approach 1:
The system performs preliminary actions by monitoring retention characteristics and threshold voltage changes before severe performance degradation occurs. The memory controller proactively identifies pages with deteriorating retention and schedules data relocation during appropriate windows, preventing the need for reactive block-level relocation only when failures occur.
Solution Approach 2:
The invention segments the data recovery process into page-level operations rather than requiring full block-level relocation. By identifying and relocating only specific pages with retention issues within a block, the system reduces the scope and time of recovery operations while maintaining overall reliability.
2Reliability
If conservative data relocation criteria are used, then data integrity is maintained, but performance degradation persists longer
Solution Approach 1:
The system dynamically adjusts data relocation criteria based on monitored retention characteristics and threshold voltage distributions. Rather than using fixed conservative thresholds, the memory controller adapts its relocation decisions to the actual degradation state of each page, enabling timely intervention before performance becomes severe while avoiding unnecessary relocation of healthy data.
Solution Approach 2:
The memory controller continuously monitors retention characteristics and threshold voltage changes, using this feedback to make informed decisions about when data relocation is necessary. This feedback mechanism allows the system to maintain data integrity while avoiding premature or unnecessary relocation operations that would reduce productivity.
3Reliability
If error bit recovery algorithm is performed excessively, then data integrity is maintained, but read performance degrades severely
Solution Approach 1:
The system performs preliminary actions by monitoring retention characteristics and threshold voltage changes before severe performance degradation occurs. The memory controller proactively identifies pages with deteriorating retention and schedules data relocation during appropriate windows, preventing the need for reactive block-level relocation only when failures occur.
Solution Approach 2:
The invention extracts the problematic pages with retention issues from the healthy pages within the same block. By isolating and relocating only the affected pages rather than treating the entire block uniformly, the system maintains data integrity for critical data while minimizing the impact on overall read performance.
Data Source
AI summary
A storage device may include a plurality of memory cells arranged in pages and blocks, each page including a row of memory cells, and each block including a plurality of pages of memory cells. The storage device may include a memory device, such as a nonvolatile memory device, which includes these items. A data recovery method for the storage device may include receiving by the storage device a first command corresponding to a first selected data recovery scheme. Based on the first command: a first target page scheme for performing error detection on the plurality of blocks is applied, target pages are read using the first target page scheme, and an amount of errors in each read target page is detected. In addition, it may be determined that a target page of a first block has at least a first threshold amount of errors, and based on the determination, data recovery for the first block may be performed by relocating all data stored in the first block to another block. Further, it may be determined that no read page in a second block has at least the first threshold amount of errors, and as a result, the data stored in the second block can be maintained.