Storage Memory Garbage Collection Using Wear-Aware Unit Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing garbage collection methods in storage devices prioritize valid data amount, leading to inefficient maintenance and reduced service life due to frequent operations on small data portions, which complicates device management and shortens the device's lifespan.
Innovation Solution
A memory management method that considers erase count, read count, and valid count information to determine a source unit for data migration, balancing execution and maintenance performance by dynamically adjusting the calculation weights of these counts to select an optimal source unit for garbage collection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If entity units storing relatively less valid data are preferentially selected as source units for garbage collection, then the execution performance of garbage collection operations is improved, but the maintenance management performance deteriorates and service life is shortened
Solution Approach 1:
The patent changes the selection parameters for source units from solely considering valid data amount to a comprehensive evaluation system that includes valid data amount, erase count, read count, and other parameters. This multi-parameter approach allows the system to balance execution performance with device reliability and service life.
Solution Approach 2:
The patent implements a feedback mechanism where the system continuously monitors multiple parameters (erase count, read count, valid data amount) of entity units and uses this information to dynamically adjust source unit selection. This feedback loop enables the system to adapt to the actual wear state and performance characteristics of the storage device.
2Productivity
If entity units storing relatively less valid data are preferentially selected as source units, then garbage collection efficiency is improved, but device management complexity increases
Solution Approach 1:
The patent transforms the simple selection criterion (valid data amount only) into a comprehensive parameter set including erase count, read count, and valid data amount. This allows the system to maintain high garbage collection efficiency while avoiding excessive management complexity through structured parameter evaluation.
Data Source
AI summary
A memory management method and a storage device are provided. The method includes: according to erase count information, read count information, and valid count information corresponding to each entity unit, obtaining quantified evaluation information corresponding to each entity unit; determining a source unit from the plurality of entity units according to the quantified evaluation information; and performing data migration on valid data in the source unit. Therefore, both the execution performance of a garbage collection operation and the maintenance management performance of the storage device may be balanced, so that the service life of the storage device may be increased.