Dynamic Refresh Rate Control for Non-Volatile Memory Cells

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Non-volatile memory devices face challenges in efficiently managing refresh operations to prevent data loss due to inadequate refresh rates, especially under varying workloads and access patterns, leading to increased write amplification and potential data failure.

Innovation Solution

A controller is configured to determine and adjust the refresh rate of memory cells based on access thresholds, miss rates, and workload types, dynamically or statically, to reduce write amplification and protect memory cells from heavy access loads by refreshing cells that have been accessed a certain number of times or are at risk of failure.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If refresh operations are performed frequently to prevent data loss, then data retention reliability is improved, but write amplification increases and device lifespan decreases

Engineering Contradiction:

Improvedata retention reliabilityVSAvoidwrite amplification

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent implements dynamic refresh rate adjustment where the controller monitors access patterns and workload types to determine optimal refresh intervals. Instead of using a fixed refresh rate, the system adapts the refresh frequency based on real-time conditions, performing more frequent refreshes when data loss risk is high and reducing refresh frequency when data is stable, thereby balancing reliability improvement with write amplification reduction.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the refresh rate parameter dynamically based on monitored conditions including access patterns, workload types, and data age. The controller adjusts this critical parameter to optimize between data retention reliability and write amplification, using multiple parameters (access count, time since last access, data importance) to determine the appropriate refresh frequency for different data blocks.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If refresh rate is increased to protect against data loss, then miss rate decreases, but energy consumption and device complexity increase

Engineering Contradiction:

Improvemiss rateVSAvoidenergy consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent dynamically adjusts the refresh rate parameter based on monitored conditions such as access patterns, workload types, and data age. By changing this parameter adaptively rather than using a fixed high refresh rate, the system achieves low miss rates when necessary while conserving energy during periods of low risk, thus resolving the contradiction between reliability and energy consumption.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system implements dynamic refresh rate adjustment where the controller continuously monitors system state and adapts refresh frequency accordingly. This dynamic approach ensures low miss rates through increased refresh activity when needed while reducing energy consumption by lowering refresh frequency when data loss risk is minimal, optimizing the trade-off between reliability and energy usage.

Inventive Principle:

Principle #15Dynamics

3Device complexity

If static refresh rate is used, then device complexity is reduced, but adaptability to varying workloads decreases leading to higher miss rates

Engineering Contradiction:

Improverefresh control complexityVSAvoidworkload adaptability

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent transitions from static to dynamic refresh rate control, where the controller monitors access patterns, workload types, and data characteristics to adaptively determine refresh intervals. This dynamic approach provides high adaptability to varying workloads while maintaining manageable device complexity through systematic monitoring and decision-making algorithms.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the refresh rate parameter based on monitored workload conditions and data characteristics. By implementing parameter changes driven by accessible monitoring data rather than complex predictive models, the patent achieves high workload adaptability while keeping device complexity at acceptable levels.

Inventive Principle:

Principle #35Parameter changes

4Productivity

If refresh operations are performed based on access thresholds, then write amplification is reduced, but data loss risk increases under heavy access loads

Engineering Contradiction:

Improvewrite amplification reductionVSAvoiddata loss prevention

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent dynamically adjusts the refresh rate parameter based on monitored conditions including access patterns, workload types, and data age. By changing this parameter adaptively rather than using a fixed high refresh rate, the system achieves low miss rates when necessary while conserving energy during periods of low risk, thus resolving the contradiction between reliability and energy consumption.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS10431286B2Refresh in non-volatile memory

Publication Date: 2019.10.01 MICRON TECHNOLOGY INC

AI summary

The present disclosure includes apparatuses and methods related to refresh in memory. An example apparatus can refresh a memory cell of an array of memory cells in response to the array of memory cells being accessed a threshold number of accesses.