Memory Subsystem Endurance Grouping for Data Retention

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing memory sub-systems lack efficient data storage management, leading to increased write amplitude and resource wastage due to the use of high endurance memory components for ephemeral data and the application of a one-size-fits-all approach to data storage.

Innovation Solution

A memory controller that coordinates with the host to select the appropriate memory component category based on data type, allowing for efficient storage by utilizing lower endurance memory components for ephemeral data and optimizing storage resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If high endurance memory components are used for all data storage, then data reliability is improved, but write amplitude increases and memory resources are wasted

Engineering Contradiction:
Improvedata reliabilityVSAvoidwrite amplitude
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent applies local quality by categorizing memory components into different endurance groups (first, second, and third groups with decreasing endurance capabilities) and assigning data to specific groups based on their retention requirements. Frequently accessed data is stored in high-endurance memory groups, while infrequently accessed data is stored in lower-endurance memory groups, ensuring each data type receives appropriate storage quality without uniform over-provisioning.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the parameter of memory component selection by dynamically choosing different memory endurance groups based on data characteristics and access patterns. The system monitors data access frequency and adjusts the placement of data across different endurance groups, changing the effective storage parameters to optimize between reliability and write amplitude reduction.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If high endurance memory components are used for all data storage, then data reliability is improved, but memory resources are wasted

Engineering Contradiction:
Improvedata reliabilityVSAvoidmemory resources
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent implements local quality by creating distinct memory storage regions with different endurance characteristics and assigning data to specific regions based on their retention needs. Critical data requiring long-term retention is placed in high-endurance memory groups, while temporary or frequently rewritten data is placed in lower-endurance groups, thereby conserving high-endurance memory resources.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent segments the memory system into multiple endurance groups (first, second, and third groups) with different reliability characteristics. This segmentation allows the system to allocate memory resources efficiently by matching data retention requirements with appropriate memory segments, preventing the waste of high-endurance resources on data that does not require them.

Inventive Principle:
Principle #1Segmentation

3Device complexity

If a one-size-fits-all approach is used for data storage, then device complexity is reduced, but memory system performance deteriorates

Engineering Contradiction:
Improvestorage management complexityVSAvoidmemory system performance
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent applies dynamics by implementing an adaptive data placement system that dynamically adjusts data storage locations based on access patterns and data characteristics. The system continuously monitors data usage and repositions data between different endurance groups as needed, allowing the storage management to evolve and optimize performance without requiring complex manual configuration.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent enables self-service through automated data classification and placement mechanisms that operate without external intervention. The memory controller automatically analyzes data access patterns, determines appropriate endurance groups, and manages data migration between groups, reducing the need for complex external storage management while maintaining optimal performance.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS20250036303A1Reclaim unit organization by endurance capability
Publication Date: 2025.01.30 MICRON TECHNOLOGY INC
  • US20250036303A1 patent drawing
  • US20250036303A1 patent drawing
  • US20250036303A1 patent drawing

AI summary

Aspects of the present disclosure configure a memory sub-system controller to allow a host to control storage on the memory sub-system based on endurance of memory components. The controller groups the set of memory components into a plurality of categories representing different endurance levels of the set of memory components and communicates, to a host, information about the plurality of categories. The controller receives, from the host, a request to program data into an individual memory component of the set of memory components, the request being generated by the host based on a type of the data and an individual category associated with the individual memory component.