Wear Leveling for Non-Volatile Memory Using Three-Pool Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wear leveling algorithms for non-volatile memory, such as the Dual-Pool algorithm, increase additional wear and slow down the search speed as storage capacity increases, due to the need to manage and balance erase operations across a growing number of physical blocks.
Innovation Solution
A wear leveling method that divides physical blocks into cool, normal, and hot pools based on their erase numbers, with dynamic threshold adjustment and the use of blank and garbage block chained lists to manage allocation and recycling, reducing additional wear and improving search efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of stationary object
If wear leveling operations are performed to balance erase numbers across physical blocks, then the service life of non-volatile memory is extended, but additional wear is caused by the wear leveling algorithms themselves
Solution Approach 1:
The patent changes the parameter of pool classification from traditional two-pool (hot/cool) to three-pool (hot/normal/cool) based on erase number thresholds. By introducing a normal pool between hot and cool pools, the system reduces unnecessary data migration operations, thereby reducing additional wear while maintaining wear leveling effectiveness
Solution Approach 2:
The patent segments the physical blocks into three distinct pools (hot pool, normal pool, cool pool) based on their erase numbers. This segmentation allows for more granular control over wear leveling operations, reducing unnecessary migrations and minimizing additional wear caused by the wear leveling process itself
2Quantity of substance
If the number of physical blocks is increased to increase storage capacity, then the storage capacity is improved, but the search speed for wear information becomes slower
Solution Approach 1:
The patent segments physical blocks into three pools (hot, normal, cool) based on erase number thresholds. This segmentation organizes the large number of physical blocks into manageable groups, improving search efficiency by reducing the scope of wear information retrieval to only relevant pools rather than searching all blocks
Solution Approach 2:
The patent applies local quality by treating different pools with different wear leveling strategies. Hot pool blocks undergo frequent wear leveling operations, normal pool blocks have moderate operations, and cool pool blocks have minimal operations. This localized approach improves overall system efficiency while maintaining wear balance
Data Source
AI summary
The present invention discloses a wear leveling method; the method determines a pool mask for each physical block based on an erase number of each physical block. For different erase numbers, masks of the physical blocks are determined as cool pool mask CPM, normal pool mask NPM or hot pool mask HPM. When the pool mask of one physical block is changed from NPM to HPM, data of any physical block of which the pool mask is CPM is copied to the physical block of which the pool mask is HPM, and the physical block of which the pool mask is CPM is recycled as a garbage block. The present invention discloses a wear leveling apparatus, the method and apparatus can reduce additional wear caused by the wear leveling.