Non-Volatile Memory Wear Leveling via Error Monitoring

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Non-volatile solid state memory devices face limitations in endurance due to finite write/erase cycles and data retention challenges, with existing wear-leveling techniques relying on cycle counters that do not accurately predict end-of-life and do not account for error rates effectively.

Innovation Solution

A method and device for managing non-volatile solid state memory by monitoring error occurrences during read operations, using error-correcting codes to identify and balance write/erase cycles, and dynamically adjusting memory programming based on error rates and thresholds to extend endurance and prevent data loss.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional wear-leveling techniques using cycle counters are used, then write/erase cycles are balanced among blocks, but the accuracy of predicting end-of-life and accounting for error rates deteriorates

Engineering Contradiction:
Improveaccuracy of predicting end-of-lifeVSAvoidaccuracy of predicting end-of-life
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent implements feedback by monitoring actual error rates during read operations and using this information to dynamically adjust wear-leveling strategies. The system continuously collects error data, analyzes trends, and modifies programming decisions based on real-time performance metrics, thereby improving accuracy in predicting end-of-life without sacrificing reliability

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent replaces the mechanical cycle counting approach with an error-rate monitoring and analysis system. Instead of relying solely on write/erase cycle counters, the system uses error detection and correction mechanisms to assess block health, substituting a more accurate software-based monitoring approach for the traditional hardware-based counting method

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If error monitoring and dynamic programming adjustment are implemented, then endurance is extended and data loss is prevented, but device complexity increases

Engineering Contradiction:
ImproveenduranceVSAvoidcomplexity of error monitoring system
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies self-service by enabling the memory system to automatically monitor its own error rates, analyze performance data, and adjust programming operations without external intervention. The system serves itself by implementing wear-leveling and error correction strategies through integrated controllers that autonomously manage block usage based on observed error patterns

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent implements preliminary action by proactively monitoring error rates and adjusting programming operations before failures occur. The system analyzes error trends and preemptively redistributes data or adjusts write patterns to prevent imminent block failures, thereby extending endurance before actual data loss occurs

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS8874993B2Memory management in a non-volatile solid state memory device
Publication Date: 2014.10.28 SK HYNIX INC
  • US8874993B2 patent drawing
  • US8874993B2 patent drawing
  • US8874993B2 patent drawing

AI summary

A non-volatile solid state memory device and method for balancing write/erase cycles among blocks to level block usage. The non-volatile solid state memory device includes a memory unit having data stored therein and a controller with logic for programming the memory unit according to a monitored occurrence of an error during a read operation. The method includes monitoring an occurrence of an error during a read operation in a memory unit of the device and programming the memory unit according to the monitored occurrence of the error.