Tag-Based Wear Leveling for Non-Volatile Memory Regions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Non-volatile storage devices face data loss due to uneven wear across regions, as improved process technologies lead to differences in operation rates, causing some regions to wear faster than others, and traditional wear leveling methods may initiate too late, exceeding error correction capabilities.

Innovation Solution

A data storage device employs tag-based wear leveling by assigning health metrics to regions based on program/erase cycles, bit error rate, and programming voltage, allowing for the relocation of infrequently accessed data from 'cold' to 'hot' regions to distribute wear evenly, and adjusting programming signals as regions age to maintain data integrity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional wear leveling based on PEC count is used, then wear is distributed among regions, but regions with similar PEC counts may wear at different rates due to process technology variations

Engineering Contradiction:
Improvedata integrityVSAvoidwear assessment accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent changes the parameter used for wear assessment from PEC count to a composite health metric that includes bit error rate, read disturbance rate, and programming voltage. This allows for more accurate identification of weak regions that are actually at risk of failure, even if they have not undergone many PEC cycles. The health metric provides a more precise measurement of actual region condition rather than just usage history.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If wear leveling is performed based on PEC counts, then data relocation is triggered, but relocation may occur too late when error rate exceeds correction capability

Engineering Contradiction:
Improveerror correction capabilityVSAvoidtime until data loss
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary assessment of region health using multiple parameters (bit error rate, read disturbance rate, programming voltage) to identify weak regions before they actually fail. By monitoring these leading indicators, the system can proactively relocate data from at-risk regions to healthier regions, preventing data loss rather than reacting after errors have already exceeded correction capability.

Inventive Principle:
Principle #10Preliminary action

3Quantity of substance

If process technology is improved to reduce component sizes, then storage density increases, but differences in operation between storage device regions increase causing uneven wear

Engineering Contradiction:
Improvestorage densityVSAvoidoperational uniformity across regions
Core Design Contradiction:
Quantity of substanceVSStability of the object's composition

Solution Approach 1:

The patent applies local quality by assigning different health metrics and wear characteristics to different regions based on their actual performance. Instead of treating all regions uniformly, the system individually assesses each region's bit error rate, read disturbance rate, and programming voltage requirements. This allows tailored management of each region's health, accommodating the operational variations introduced by advanced process technologies.

Inventive Principle:
Principle #3Local quality

4Measurement precision

If health metric assessment is implemented, then accurate region selection for wear leveling is achieved, but additional monitoring and calculation overhead is introduced

Engineering Contradiction:
Improveregion health assessment accuracyVSAvoidwear leveling system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs self-service by utilizing existing operational data (bit errors during reads, read disturbance effects, programming voltages already applied during writes) to calculate health metrics. Rather than requiring separate monitoring circuits or additional sensors, the wear assessment leverages data already generated during normal storage operations, reducing the need for additional hardware complexity while maintaining high measurement precision.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS9740425B2Tag-based wear leveling for a data storage device
Publication Date: 2017.08.22 SANDISK TECHNOLOGIES LLC
  • US9740425B2 patent drawing
  • US9740425B2 patent drawing
  • US9740425B2 patent drawing

AI summary

A data storage device includes a memory. A method includes de-allocating a first region of a group of regions of the memory during a wear leveling process based on a determination that the first region is associated with a first tag of a set of tags. Each region of the group of regions is assigned to a tag of the set of tags based on a health metric associated with the region. The health metric is based on a bit error rate (BER), a program/erase cycle (PEC) count, a PEC condition metric, or a combination thereof. In response to selecting the first region, information is copied from the first region to a second region of the memory during the wear leveling process.