Dynamic Garbage Collection Threshold for Memory Storage Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Rewritable non-volatile memory devices frequently perform garbage collection operations due to a fixed threshold value, leading to decreased operation performance even when there is sufficient storage space, as the number of physical programming units increases without a corresponding increase in valid data.
Innovation Solution
A memory management method that dynamically adjusts a threshold value based on the valid data amount and the number of physical erasing units programmed in single-page mode, delaying the start of garbage collection and allowing for adaptive programming mode changes between single-page and multi-page modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a fixed threshold value is used to determine when to perform garbage collection, then the memory storage device can automatically trigger garbage collection, but the frequency of garbage collection increases unnecessarily leading to reduced operation performance
Solution Approach 1:
The patent applies dynamics by changing the threshold value from fixed to dynamic. The threshold is now determined based on the valid data amount stored in physical erasing units, allowing the garbage collection trigger point to adapt to actual data usage conditions. This resolves the contradiction by enabling automatic garbage collection while preventing unnecessary frequent triggers that would reduce operation performance.
Solution Approach 2:
The patent changes the parameter used for garbage collection determination from a fixed threshold to a dynamic threshold based on valid data amount. By adjusting this parameter according to actual storage conditions, the system achieves both automation and optimal performance, avoiding the pitfall of excessive garbage collection frequency.
2Quantity of substance
If the number of physical programming units increases due to continuous write operations, then more data can be stored, but the number of physical erasing units increases without corresponding valid data increase, causing frequent garbage collection
Solution Approach 1:
The patent implements feedback by continuously monitoring the valid data amount stored in physical erasing units and using this information to determine the threshold for garbage collection. This feedback mechanism allows the system to distinguish between actual storage needs and artificial inflation of physical programming unit counts, thereby maintaining high productivity while utilizing storage capacity effectively.
Solution Approach 2:
The patent replaces the mechanical counting of physical programming units with a more intelligent system that measures valid data amount. This substitution allows the system to ignore the quantity of physical units and focus on actual useful data stored, preventing false triggers of garbage collection and maintaining optimal operation performance.
3Reliability
If garbage collection is performed frequently to manage physical erasing units, then data validity is maintained, but the overall operation performance is influenced due to excessive collection frequency
Solution Approach 1:
The patent changes the parameter for garbage collection determination from fixed to dynamic based on valid data amount. This ensures that garbage collection is performed only when actually needed to maintain data validity, rather than following a rigid schedule that would reduce operation performance. The dynamic threshold adapts to real storage conditions, balancing reliability and productivity.
Solution Approach 2:
The patent applies dynamics by making the garbage collection threshold adaptive rather than static. The threshold dynamically adjusts according to the valid data amount, allowing the system to maintain data validity through intelligent timing of garbage collection while avoiding excessive frequency that would harm operation performance.
Data Source
AI summary
A memory management method is provided according to an exemplary embodiment of the disclosure. The method includes: obtaining a valid data parameter based on a valid data amount of valid data stored in a plurality of physical erasing units, and obtaining a first threshold value based on the valid data parameter. The method also includes: obtaining a first determination parameter based on a number of a plurality of first physical erasing units, and the first physical erasing units are physical erasing units being programmed for storing data by using a single-page programming mode. The method further includes: performing a garbage collection operation if the first determination parameter is greater than the first threshold value.


