Memory Refresh Module Reduces Voltage Shifts Without Erasure

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data retention issues in memory devices, such as flash memory, lead to degradation of multi-level cells, causing errors in data reading due to voltage shifts, which affect system performance and require frequent reprogramming, resulting in write amplification and reduced program-erase cycling capability.

Innovation Solution

A memory device with integrated processor circuitry that includes an erase module, program module, and refresh module to refresh data without erasure, and a memory controller with error correction and refresh trigger modules to determine and execute data refresh based on error detection, using a different programming scheme with fewer steps than initial programming.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If data is programmed into multi-level cells for higher storage capacity, then storage density is improved, but data retention deteriorates due to voltage shifts and degradation

Engineering Contradiction:
Improvestorage densityVSAvoiddata retention
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent applies preliminary action by performing a refresh operation before data is completely degraded. The system monitors data integrity and executes a refresh programming operation to restore voltage levels before significant degradation occurs, preventing read errors and maintaining data retention in multi-level cells

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes programming parameters by using a simplified programming scheme for refresh operations compared to initial programming. The refresh operation uses different voltage levels and fewer programming steps optimized for restoring data in multi-level cells, thereby improving data retention while maintaining storage density

Inventive Principle:
Principle #35Parameter changes

2Reliability

If frequent reprogramming is performed to maintain data integrity, then data retention is improved, but write amplification increases and program-erase cycling capability decreases

Engineering Contradiction:
Improvedata retentionVSAvoidprogram-erase cycling capability
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies local quality by performing refresh operations selectively on only those data blocks or pages that show signs of degradation, rather than refreshing the entire memory. This targeted approach maintains data retention for affected regions while minimizing unnecessary write operations on already healthy data, thereby reducing write amplification and preserving program-erase cycling capability

Inventive Principle:
Principle #3Local quality

3Speed

If a simplified programming scheme is used for refresh operations, then programming speed is improved, but programming precision may deteriorate

Engineering Contradiction:
Improveprogramming speedVSAvoidprogramming precision
Core Design Contradiction:
SpeedVSManufacturing precision

Solution Approach 1:

The patent applies partial action by using a simplified programming scheme that performs only the essential programming steps needed for refresh operations, omitting certain verification and optimization steps used in initial programming. This achieves sufficient precision for restoring data while significantly improving programming speed for refresh operations

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS10353598B2System and method for refreshing data in a memory device
Publication Date: 2019.07.16 SANDISK TECHNOLOGIES LLC
  • US10353598B2 patent drawing
  • US10353598B2 patent drawing
  • US10353598B2 patent drawing

AI summary

Systems, apparatuses, and methods are provided that refresh data in a memory. Data is programmed into the memory. After which, part or all of the data may be refreshed. The refresh of the data may be different from the initial programming of the data in one or more respects. For example, the refresh of the data may include fewer steps than the programming of the data and may be performed without erasing a section of memory. Further, the refresh of the data may be triggered in one of several ways. For example, after programming the data, the data may be analyzed for errors. Based on the number of errors found, the data may be refreshed.