SSD Buffer Preservation Using Dynamic Fill Ratio Thresholds

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Semiconductor storage devices face performance degradation due to limited single-level cell (SLC) blocks, as QLC blocks are configured to function as SLC blocks, reducing the number of available SLC blocks and degrading write and read performance when fill ratio exceeds a certain threshold.

Innovation Solution

Implement a dynamic data preservation (DDP) threshold policy that dynamically adjusts the threshold value based on fill ratio, using a machine learning model to identify important data and migrate it to low-performance buffer blocks during idle time, maintaining an optimal balance between read and write performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If QLC blocks are configured to function as SLC blocks to increase the number of high-performance blocks, then the capacity for high-performance storage increases, but the fill ratio increases and degrades write and read performance when the threshold is exceeded

Engineering Contradiction:
Improvenumber of high-performance blocksVSAvoidwrite and read performance
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent implements a dynamic threshold policy that adjusts the fill ratio threshold based on real-time monitoring of HPB block usage. When the fill ratio exceeds a predefined preservation threshold, the system dynamically reconfigures QLC blocks to function as SLC blocks, increasing the number of high-performance blocks. This dynamic adjustment resolves the contradiction by adapting the system configuration based on current workload conditions, ensuring optimal performance while maximizing high-performance storage capacity.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the operational parameters of memory blocks by reconfiguring QLC blocks to function as SLC blocks when performance degradation is detected. This parameter change transforms the storage characteristics of the blocks, converting them from high-capacity low-performance mode to high-performance mode, thereby resolving the contradiction between quantity and productivity.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If the fill ratio threshold is set low to maintain performance, then write and read operations remain fast, but the capacity utilization of high-performance buffer blocks is reduced

Engineering Contradiction:
Improvewrite and read speedVSAvoidcapacity utilization
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent employs a feedback mechanism that continuously monitors the fill ratio of HPB blocks and performance metrics. Based on this feedback, the system dynamically adjusts the threshold for reconfiguring QLC blocks to SLC blocks. This feedback loop allows the system to maintain high performance by keeping the fill ratio within acceptable ranges while maximizing capacity utilization, resolving the contradiction between speed and capacity utilization.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs self-adjustment by automatically reconfiguring blocks and managing data migration based on monitored conditions without external intervention. This self-service capability enables the system to maintain optimal performance while maximizing capacity utilization through automatic threshold adjustment and block reconfiguration.

Inventive Principle:
Principle #25Self-service

3Speed

If more QLC blocks are used as SLC blocks to improve write performance, then write speed increases, but the number of available QLC blocks for capacity storage decreases

Engineering Contradiction:
Improvewrite speedVSAvoidavailable capacity storage
Core Design Contradiction:
SpeedVSQuantity of substance

Solution Approach 1:

The patent implements dynamic reconfiguration of QLC blocks to SLC blocks based on real-time performance monitoring and fill ratio thresholds. This dynamic approach allows the system to increase write speed by converting QLC blocks to SLC blocks only when necessary, rather than permanently allocating blocks, thereby maintaining capacity storage availability while improving write performance when needed.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the operational parameters of QLC blocks dynamically, converting them to SLC mode temporarily to improve write speed, and can revert them when performance requirements change. This parameter flexibility resolves the contradiction between write speed and capacity storage by allowing the same physical blocks to serve different functions based on current system needs.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4675415A1Apparatus for providing dynamic data preservation in a storage device and operating method thereof
Publication Date: 2026.01.07 SAMSUNG ELECTRONICS CO LTD
  • EP4675415A1 patent drawingFigure 1
  • EP4675415A1 patent drawingFigure 2
  • EP4675415A1 patent drawingFigure 3

AI summary

A method of providing dynamic data preservation in a storage device and an apparatus (106) configured to perform the method are provided. The apparatus (106) provides dynamic data preservation in a storage device (104), in which the apparatus includes control module (208) configured to monitor (702) a fill ratio of the storage device (104), compare (704) the fill ratio with a predefined preservation fill ratio of the storage device to determine whether the fill ratio exceeds the predefined preservation fill ratio, dynamically configure a dynamic data preservation, DDP, threshold value of the storage device (104) based on the fill ratio of the storage device (104) when the fill ratio exceeds the predefined preservation fill ratio, and identify (708) important data among data stored in high performance buffer, HPB, blocks of the storage device (104) and migrate remaining data which are not identified as important data to low performance buffer, LPB, blocks of the storage device (104) based on the configured DDP threshold value during idle time of the storage device (104). Alternatively, a dynamic data preservation, DDP, threshold value of the storage device (104) may be determined based on the fill ratio using a machine learning model (606).