Memory Controller Read Threshold Optimization via Asymmetric Ratio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current memory systems face inefficiencies in read operations due to high read latency and quality of service degradation caused by multiple read retry operations, particularly in triple level cell (TLC) and quadruple level cell (QLC) memory devices, which increase read latency and reduce performance.
Innovation Solution
A memory system and method that optimize read threshold values using domain transformation by determining an asymmetric ratio (AR) and number of unsatisfied checks (USCs) for decoded data, arranging AR values along a Z-axis, and estimating optimum read threshold sets based on coordinate values corresponding to specific AR and USC values, thereby reducing read latency and improving quality of service.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple read retry operations are performed to improve data reading accuracy, then reading accuracy is improved, but read latency increases and quality of service degrades
Solution Approach 1:
The patent changes the parameter space by transforming read threshold values into asymmetric ratio (AR) values through a monotonic transformation function. This allows the system to work with AR values that have a more favorable distribution characteristic, enabling faster convergence to optimal thresholds with fewer read retry operations, thus reducing read latency while maintaining reading accuracy.
Solution Approach 2:
The patent performs preliminary transformation of read threshold values to asymmetric ratio values before the actual read operations. By pre-establishing the AR value domain and using it to guide the read threshold optimization process, the system avoids unnecessary read retry operations, thereby reducing read latency while ensuring accurate data reading.
2Measurement precision
If traditional read threshold optimization methods (e.g., eBoost) are used to ensure accurate data reading, then reading accuracy is maintained, but numerous read retry operations are required increasing operational complexity
Solution Approach 1:
The patent transforms the read threshold parameter space into an asymmetric ratio parameter space using a monotonic transformation function. This parameter change simplifies the optimization process by creating a more favorable distribution characteristic, allowing the system to achieve accurate reading with fewer read retry operations and reduced operational complexity compared to traditional methods like eBoost.
Solution Approach 2:
The patent replaces the mechanical iterative read retry process with a domain transformation approach. Instead of repeatedly performing read operations with different thresholds, the system transforms thresholds to AR values and uses coordinate values on a Z-axis to determine optimal thresholds, substituting a computational mathematical approach for a procedural iterative approach, thereby reducing operational complexity.
3Loss of time
If read threshold values are optimized using domain transformation to reduce read latency, then quality of service is improved, but the complexity of the optimization algorithm increases
Solution Approach 1:
The patent applies parameter transformation by converting read threshold values to asymmetric ratio values through a monotonic transformation function. This creates a more favorable distribution characteristic that enables faster convergence, reducing read latency. The algorithmic complexity is managed by establishing a systematic framework with monotonic transformation and coordinate-based optimization that provides clear computational steps.
Solution Approach 2:
The patent introduces a new dimension by transforming the one-dimensional read threshold parameter into an asymmetric ratio parameter space. This dimensional transformation creates a more favorable optimization landscape with better distribution characteristics, allowing the system to find optimal thresholds faster and reduce read latency, while the structured approach manages the increased algorithmic complexity.
Data Source
AI summary
A controller optimizes read threshold values for a memory device using domain transformation. The controller determines, for decoded data of each read operation, an asymmetric ratio (AR) and a number of unsatisfied checks (USCs), the AR indicating a ratio of a number of a first binary value to a number of a second binary value in the decoded data. The controller determines a Z-axis such that AR values of threshold sets are arranged in a set order along the Z-axis. The controller determines an optimum read threshold set using coordinate values on the Z-axis, which correspond to a set AR value and a set USC value.