Memory Block Garbage Collection Based on Variable Life Threshold
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional garbage collection operations in nonvolatile memory devices are time-consuming, reducing overall performance due to the need to invalidate and erase entire memory blocks, leading to significant unavailable memory space.
Innovation Solution
A method and system for managing memory blocks in nonvolatile memory devices that identify and select full blocks with a block life exceeding a threshold as target blocks for garbage collection, allowing young blocks sufficient time to invalidate pages before selecting old blocks for garbage collection, thereby reducing data movement and improving performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional garbage collection operations are performed on full memory blocks, then unavailable memory space is freed up, but the operation requires significant time and reduces overall device performance
Solution Approach 1:
The patent changes the selection parameter for garbage collection from arbitrary or simple criteria to block life threshold. By monitoring and comparing block life against a predefined threshold, the system identifies optimal candidates for garbage collection that balance memory space reclamation with performance preservation. This parameter-based selection reduces unnecessary garbage collection operations on recently written blocks, thereby reducing total garbage collection time while maintaining adequate memory space availability.
2Quantity of substance
If entire memory blocks are erased to free up space, then memory space is reclaimed, but significant data movement is required which increases operation time
Solution Approach 1:
The patent introduces block life threshold as a selection criterion that prevents premature garbage collection on blocks with recent writes. By filtering out young blocks (those below the threshold) from garbage collection candidates, the system avoids unnecessary data movement operations. This selective approach based on block life parameter ensures that garbage collection is performed only on blocks where data movement is truly necessary, thereby maintaining device productivity while still reclaiming memory space.
Data Source
AI summary
Memory blocks of a nonvolatile memory device are managed by identifying a full memory block, determining whether a block life of the full memory block exceeds a threshold value, and upon determining that the block life of the full memory block exceeds the threshold value, selecting the full memory block as a target block for garbage collection. The threshold of the block life is determined using an average write distance of logical pages programmed in the nonvolatile memory device.