Binary Cache Cleanup Using Fragment Age Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current binary cache eviction algorithms in flash memory management systems fail to select the most suitable logical groups for eviction, leading to reduced usable capacity, increased frequency of cache evictions, and lower performance due to higher compaction frequencies.
Innovation Solution
A method and system for binary cache cleanup that includes eviction mechanisms analyzing cache storage for predetermined criteria such as oldest data or compaction history, and selecting logical space indicators for eviction based on these analyses, with multiple eviction mechanisms used during different phases of operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the current binary cache eviction algorithm selects logical groups with the highest number of fragments for eviction, then the cache utilization threshold can be exceeded, but the usable capacity is reduced and the frequency of cache evictions increases
Solution Approach 1:
The patent changes the selection criterion from 'number of fragments' to 'age of oldest fragment'. This parameter change transforms the eviction strategy from quantity-based to time-based selection, identifying logical groups whose oldest fragment has not been accessed for the longest period. This resolves the contradiction by evicting truly obsolete data rather than merely fragmented data, thereby maintaining usable capacity while reducing unnecessary eviction frequency.
Solution Approach 2:
The system uses the age information of fragments already stored in the cache to make eviction decisions, rather than requiring external metrics or complex analysis. Each fragment carries its own age characteristic that serves as the basis for eviction selection, enabling the cache to self-regulate its contents based on inherent data characteristics.
2Reliability
If the current binary cache eviction algorithm increases the frequency of cache evictions to the MLC, then the cache can be kept cleaner, but the performance decreases and MLC endurance is reduced
Solution Approach 1:
The system performs preliminary identification of obsolete logical groups using the oldest fragment criterion before eviction is needed. By pre-selecting candidates based on fragment age, the system prepares a ready list of eviction targets that minimizes subsequent eviction operations, thereby maintaining cache cleanliness without requiring frequent disruptive evictions to the MLC.
3Stability of the object's composition
If the current binary cache eviction algorithm increases the frequency of cache compactions, then the cache structure can be maintained, but the performance decreases and binary cache endurance is reduced
Solution Approach 1:
The patent changes the eviction trigger from fragment count-based to age-based selection. This parameter change reduces the need for frequent compactions because the system evicts entire logical groups based on the age of their oldest fragment, rather than attempting to manage individual fragments through repeated compaction operations. This maintains cache structure stability while reducing performance-degrading compaction frequency.
Data Source
AI summary
A system and method for clearing data from a cache in a storage device is disclosed. The method may include analyzing the cache for the least recently fragmented logical group, and evicting the entries from the least recently fragmented logical group. Or, the method may also include analyzing compaction history and selecting entries for eviction based on the analysis of the compaction history. The method may also include scheduling of different eviction mechanisms during various operations of the storage device. The system may include a cache storage, a main storage and a controller configured to evict entries associated with a least recently fragmented logical group, configured to evict entries based on analysis of compaction history, or configured to schedule different eviction mechanisms during various operations of the storage device.