Predictive Wear Leveling for Memory Cell Endurance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Typical wear leveling schemes in memory devices assume uniform endurance among groupings of memory cells, leading to over-utilization of cells with lower endurance and under-utilization of cells with higher endurance, resulting in premature failure of lower endurance cells and reduced rated capacity of the memory.
Innovation Solution
A system that generates a predictive model using machine learning techniques to estimate the expected endurance of groupings of memory cells based on characterization and process data, allowing for more informed wear leveling decisions to extend the life of the memory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If typical wear leveling schemes are used that assume uniform endurance among memory cell groupings, then wear leveling operations can be performed with simple tracking mechanisms, but lower endurance cells are over-utilized and fail prematurely
Solution Approach 1:
The patent changes the parameter being tracked from simple program/erase cycle counts to a more comprehensive metric that incorporates cell-specific endurance characteristics. By modifying what data is collected and how wear is measured, the system achieves better reliability without requiring fundamentally new wear leveling mechanisms.
Solution Approach 2:
The system performs preliminary characterization of memory cell groupings to determine their relative endurance before normal operation begins. This advance knowledge allows the wear leveling algorithm to make informed decisions from the start, preventing premature failure of weaker cells rather than reacting to failures after they occur.
2Ease of operation
If wear leveling distributes data uniformly across all memory blocks, then implementation is straightforward, but blocks with lower endurance are over-utilized and fail earlier
Solution Approach 1:
The patent applies local quality by treating different memory blocks differently based on their individual endurance characteristics. Instead of uniform wear leveling, the system tailors the wear distribution strategy to each block's specific properties, assigning weaker blocks to less frequently used data or positioning them in the wear leveling sequence to extend their operational life.
3Productivity
If all memory cells are treated equally in wear leveling, then the algorithm is simple to implement, but rated capacity is reduced due to premature failure of weaker cells
Solution Approach 1:
The patent segments the memory system into groups based on endurance characteristics, allowing different wear leveling strategies to be applied to different segments. This segmentation enables the system to optimize for both simplicity and reliability by treating high-endurance and low-endurance blocks differently while maintaining overall system efficiency.
Data Source
AI summary
Systems might include tester hardware for connection to a die containing a memory comprising a plurality of groupings of memory cells and a predictive model in communication with the tester hardware, wherein a controller of the tester hardware is configured to generate characterization data corresponding to a first grouping of memory cells, wherein the predictive model is configured to generate an indication of expected endurance for the first grouping of memory cells in response to the characterization data in response to process data corresponding to the first grouping of memory cells, and wherein the controller is further configured to store a value to the memory indicative of the indication of expected endurance. Methods could use the predictive model in wear leveling within a memory, and memories could use information regarding expected endurance in wear leveling.