Dynamic Read Count Threshold Adjustment for Flash Memory Reliability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The challenge in flash memory storage devices is setting an appropriate read count threshold to prevent read disturb, as stringent thresholds lead to processing waste and loose thresholds result in data loss, while existing methods fail to adaptively adjust based on actual usage conditions.

Innovation Solution

A memory control method that dynamically adjusts the read count threshold for physical erasing units by monitoring error bit amounts and voltage distribution variations, using models like SVM or RNN to predict and compare voltage distributions, thereby determining the need for refresh operations and recalculating thresholds based on usage parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a more stringent read count threshold is adopted to prevent read disturb, then data reliability is improved, but processing cost increases

Engineering Contradiction:
Improvedata reliabilityVSAvoidprocessing cost
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent implements dynamic adjustment of read count thresholds based on actual memory cell conditions. Instead of using a fixed stringent threshold for all physical erasing units, the system monitors error bit amounts and voltage distribution variations in real-time, and adjusts thresholds adaptively. This allows the system to maintain high reliability when needed while avoiding unnecessary processing costs when conditions permit, thus resolving the contradiction between data reliability and processing cost.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of read count threshold from a fixed value to a dynamically adjustable value based on monitored conditions. By detecting error bit amounts and voltage distribution variations, the system modifies the threshold parameter according to actual memory cell states. This parameter change enables the system to optimize between reliability and processing cost by applying stringent thresholds only when necessary.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If a relatively loose read count threshold is adopted to reduce processing cost, then processing efficiency is improved, but data reliability deteriorates

Engineering Contradiction:
Improveprocessing efficiencyVSAvoiddata reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system dynamically adjusts read count thresholds based on monitored error bit amounts and voltage distribution variations. When memory cells are in good condition, looser thresholds are applied to improve processing efficiency. When degradation is detected, thresholds are tightened to maintain reliability. This dynamic approach resolves the contradiction by allowing the system to operate efficiently when safe and conservatively when needed.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements a feedback mechanism that continuously monitors error bit amounts and voltage distribution variations in physical erasing units. This feedback information is used to adjust read count thresholds adaptively. The feedback loop enables the system to maintain optimal balance between processing efficiency and data reliability by responding to actual memory cell conditions rather than using fixed thresholds.

Inventive Principle:
Principle #23Feedback

3Device complexity

If fixed read count thresholds are used to simplify control, then device complexity is reduced, but adaptability to usage conditions deteriorates

Engineering Contradiction:
Improvecontrol complexityVSAvoidadaptability to usage conditions
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent transitions from static fixed thresholds to dynamic adaptive thresholds that respond to actual usage conditions. The system monitors error bit amounts and voltage distribution variations and adjusts thresholds accordingly. This dynamic approach significantly improves adaptability to different usage conditions while the automated nature of the adjustment keeps control complexity manageable.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system performs self-adjustment of read count thresholds based on its own monitored conditions. The memory control circuit automatically detects error patterns and voltage variations, then autonomously modifies thresholds without requiring external intervention or complex control algorithms. This self-service approach enhances adaptability while keeping the control mechanism relatively simple.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS12147674B1Memory control method, memory storage device and memory control circuit unit
Publication Date: 2024.11.19 HEFEI CORE STORAGE ELECTRONICS LTD
  • US12147674B1 patent drawing
  • US12147674B1 patent drawing
  • US12147674B1 patent drawing

AI summary

A memory control method, a memory storage device and a memory control circuit unit are provided. The method includes: setting preset read count thresholds corresponding to physical erasing units respectively; in a background operation, in response to a read count of a first physical erasing unit in the physical erasing units being greater than its corresponding preset read count threshold, reading word lines in the first physical erasing unit to obtain first error bit amounts; determining whether a refresh operation needs to be performed on the first physical erasing unit according to first error bit amounts; in response to no need to perform the refresh operation on the first physical erasing unit, selecting a first word line with the largest first error bit amount in the word lines, and detecting a voltage distribution variation of the first word line; and calculating a new read count threshold of the first physical erasing unit according to the voltage distribution variation.