Memory Subsystem Read Level Adaptation via Voltage Distribution Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory devices face increased bit error rates due to temporal voltage shift caused by slow charge loss, which affects the accuracy of read operations in non-volatile memory systems like NAND devices.
Innovation Solution
A memory sub-system controller determines read level values based on voltage distribution parameters to mitigate temporal voltage shift by measuring and storing distribution voltages for blocks or block families, using a voltage mapping table to identify suitable read levels for accurate data retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional fixed read levels are used, then device complexity is reduced, but read error rate increases due to temporal voltage shift
Solution Approach 1:
The system performs preliminary calibration to determine voltage distribution parameters (mean, standard deviation) for each block family before actual read operations. These parameters are stored and reused for multiple reads, avoiding repeated complex measurements while maintaining high read accuracy despite temporal voltage shift
Solution Approach 2:
The system dynamically adjusts read levels by calculating offsets based on voltage distribution parameters specific to each block family's age and wear characteristics. Instead of using fixed read levels, the system modifies read level parameters adaptively to compensate for temporal voltage shift, thereby reducing read errors without requiring complex real-time measurement for every read operation
2Measurement precision
If per-block voltage distribution measurement is performed, then read accuracy is improved, but storage space and operation time increase
Solution Approach 1:
The system merges voltage distribution measurements across multiple blocks with similar characteristics (same block family, similar age and wear) to determine representative voltage distribution parameters. By combining measurements from multiple blocks rather than treating each block independently, the system achieves sufficient measurement precision while reducing the total number of measurements required, thereby saving time and resources
3Measurement precision
If per-block voltage distribution measurement is performed, then read accuracy is improved, but device complexity increases
Solution Approach 1:
The system segments blocks into block families based on similar characteristics (age, wear, programming patterns) and determines voltage distribution parameters at the block family level rather than individual block level. This segmentation reduces the complexity of tracking voltage distributions while maintaining sufficient measurement precision through group-based characterization
Solution Approach 2:
The system creates representative voltage distribution parameter sets for block families that can be copied and applied to multiple blocks within the family. Instead of maintaining separate complex voltage distribution models for each block, the system uses copied parameter sets from representative blocks, reducing overall system complexity while preserving read accuracy
Data Source
AI summary
A method can include receiving a request to read data from a block of a memory device, identifying a block family associated with the block of the memory device, identifying a voltage distribution parameter value associated with the block family, wherein the voltage distribution parameter value reflects an aggregate value of a corresponding voltage distribution associated with a plurality of memory cells of the block family, and determining a set of read levels associated with the voltage distribution parameter value, wherein each read level in the set of read levels corresponds to a respective voltage distribution of at least one memory cell comprised by the block of the memory device. The block family can be identified using a data structure that maps block identifiers to corresponding block family identifiers. The voltage distribution parameter value can be identified using a data structure that maps block family identifiers to corresponding voltage parameter values.