Read Retry Voltage Ranking for Memory Page Populations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Solid state storage devices face challenges in maintaining accurate read threshold voltages over time due to memory cell leakage and disturbances, leading to data errors and the need for improved read retry operations.
Innovation Solution
The method involves ranking read reference voltages for different page populations based on performance metrics like bit error rate and using these rankings to perform read retry operations, with separate channel estimation for each population to account for variations, and dynamically calculating optimal read reference voltages for failing pages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple read retry operations are performed with different read reference voltages, then the bit error rate is reduced and decoding success improves, but the system latency and read time increase
Solution Approach 1:
The patent pre-calculates and stores optimal read reference voltages for different page populations and channel conditions before actual read operations. When a read retry is needed, the system quickly retrieves the pre-determined optimal voltage from lookup tables based on population classification, avoiding time-consuming real-time voltage optimization while still achieving low bit error rates
Solution Approach 2:
The patent divides the memory page population into multiple sub-populations based on channel conditions and characteristics. Each sub-population has its own pre-optimized read reference voltage settings. This segmentation allows the system to select appropriate voltage settings more quickly than optimizing for all possible conditions, reducing latency while maintaining reliability
2Measurement precision
If read reference voltages are optimized for each page population, then the quality of soft decisions improves, but the device complexity and processing overhead increase
Solution Approach 1:
The patent creates simplified models or lookup tables that copy the essential characteristics of optimal read reference voltage settings for different page populations. Instead of performing complex real-time optimization, the system uses pre-computed voltage settings stored in lookup tables, which replicate the效果 of detailed optimization with much lower processing overhead
Solution Approach 2:
The patent changes the approach from continuous real-time voltage optimization to discrete pre-selected voltage levels. By quantizing the read reference voltage into specific predetermined levels for different population types, the system achieves high soft decision quality while reducing the complexity of voltage control and processing
3Measurement precision
If separate channel estimation is performed for each page population, then the accuracy of read voltage selection improves, but the computation time and processing resources increase
Solution Approach 1:
The patent performs channel estimation and determines optimal read reference voltages for different page populations in advance, during idle periods or initial system operation. These pre-computed channel characteristics and optimal voltage settings are stored for later use, eliminating the need for time-consuming real-time channel estimation during actual read operations
Data Source
AI summary
Methods and apparatus are provided for read retry operations with read reference voltages ranked for different page populations of a memory. One method comprises obtaining a plurality of rankings of a plurality of read reference voltages for a plurality of page populations, wherein the rankings are based on a predefined performance metric; and reading a codeword from the memory a plurality of times, wherein each of the read operations uses a different one of the plurality of read reference voltages selected based on the rankings of the plurality of read reference voltages. The performance metric comprises, for example, a bit error rate, a bit polarity disparity, a substantially minimal syndrome weight and/or measures of an average system latency or a tail latency. The ranking is optionally based on a size of the page populations that had each of the ranked read reference voltages. Channel estimation is performed separately for each of the plurality of page populations.