Solid State Memory Hard Error Recovery with Modified Read Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Solid state memory devices face data loss due to cell degradation, leading to threshold voltage shifts and read errors, which conventional methods like re-reading or erasing fail to recover effectively, especially with increasing data density and error correction code decoding failures.
Innovation Solution
A method is introduced to identify hard errors in solid state memory cells, provide their locations to a decoder, and use modified reference signals and erasure flags for effective data recovery, including reprogramming with known patterns and bit-flipping techniques to correct errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional methods (re-reading or erasing) are used to recover data from degraded cells, then the process is simple and fast, but data recovery fails when hard errors occur due to cell degradation
Solution Approach 1:
The system performs preliminary identification of hard errors in memory cells before attempting data recovery. By detecting cells with threshold voltage shifts or write-in errors in advance, the system can prepare appropriate recovery strategies (such as using alternative read methods or error correction codes) rather than simply retrying conventional reads that are guaranteed to fail.
Solution Approach 2:
The invention changes the reading parameters by using modified reference signals instead of normal reference signals when reading from degraded cells. This parameter change allows the read operation to accommodate cells with shifted threshold voltages, enabling successful data retrieval from cells that would otherwise be considered unreadable by conventional methods.
2Reliability
If multiple reread and ECC decoding operations are performed on degraded cells, then more error correction attempts are made, but write-in errors cause decoding failure with little improvement
Solution Approach 1:
The system performs preliminary identification of hard errors to distinguish between soft errors (correctable by ECC) and hard errors (write-in errors or severe degradation). This preliminary classification prevents wasteful repeated ECC decoding attempts on cells with write-in errors, as the system can immediately identify these cells and apply alternative recovery methods or mark them for erasure.
Solution Approach 2:
The error correction system serves itself by automatically identifying hard errors and routing them to appropriate handling procedures without requiring multiple blind retry attempts. The system self-adapts its error correction strategy based on the detected error type, reducing unnecessary operations and time consumption.
3Quantity of substance
If data density increases in solid state memory devices, then storage capacity improves, but error correction code decoding becomes less effective due to higher error rates
Solution Approach 1:
The error correction approach is segmented into multiple stages: first identifying hard errors, then handling soft errors with ECC decoding. This segmentation allows the system to apply different error correction strategies to different types of errors, improving overall effectiveness in high-density storage where both hard and soft errors may occur.
Solution Approach 2:
By performing preliminary hard error identification before ECC decoding, the system removes or flags cells with write-in errors that would otherwise cause ECC decoding failures. This preliminary action cleans up the data set, allowing ECC decoding to focus on correctable soft errors and improving overall decoding effectiveness in high-density storage environments.
Data Source
AI summary
Embodiments herein provide data recovery techniques and configurations for solid state memory devices. For example, a method includes identifying a hard error associated with a cell of a solid state memory device, providing a location of the cell having the identified hard error to a decoder to recover data originally programmed to the cell, and recovering the data originally programmed to the cell using the decoder. Other embodiments may be described and/or claimed.