Cache Scrubber Logic for Dynamic Data Eviction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cache memory systems face data loss due to power failures or processor errors, as they rely on volatile memory and lack effective mechanisms to ensure timely transfer of persistent data to non-volatile memory, leading to potential file corruption.
Innovation Solution
Implementing a scrubber logic controller within the cache device that dynamically adjusts the aggressiveness of data eviction based on time constraints, using criteria such as data age and availability, to ensure persistent modified data is transferred to non-volatile memory within a reasonable time frame, thereby minimizing data loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If volatile cache memory is used to maintain performance, then access speed is improved, but data reliability deteriorates due to potential data loss from power failure or processor errors
Solution Approach 1:
The scrubber logic performs preliminary actions by continuously monitoring data age in the cache and proactively evicting persistent modified data before a failure event occurs. This advance action ensures that data is transferred to non-volatile memory in advance, preventing data loss without requiring backup batteries or other reactive measures.
Solution Approach 2:
The scrubber logic implements a feedback mechanism by continuously monitoring the age of data in the cache and adjusting eviction operations based on this information. When data exceeds a predetermined age threshold, the scrubber logic triggers eviction to non-volatile memory, creating a closed-loop control system that maintains data reliability while preserving cache performance.
2Productivity
If data is held in volatile cache for extended periods, then performance is improved, but the risk of data loss increases
Solution Approach 1:
The scrubber logic introduces dynamic behavior to the cache system by continuously adjusting data eviction based on real-time conditions. The logic dynamically monitors data age and adjusts eviction timing, allowing the system to optimize between performance and reliability rather than using static eviction policies. This dynamic approach ensures persistent modified data is evicted at appropriate intervals without unduly impacting performance.
3Reliability
If aggressive data eviction is performed to ensure data safety, then data reliability is improved, but cache performance deteriorates due to increased eviction overhead
Solution Approach 1:
The scrubber logic applies local quality by selectively evicting only persistent modified data that meets specific criteria (data age threshold), rather than uniformly evicting all cache data. This targeted approach ensures data safety for persistent modified data while minimizing unnecessary evictions that would degrade performance. The logic treats different data differently based on its persistence requirements and age.
Data Source
AI summary
Embodiments of the present disclosure are directed towards a computing device having a cache memory device with a scrubber logic. In some embodiments, the scrubber logic controller may be coupled with the cache device, and may perform a selection for eviction from the cache device a portion of data stored in the cache device, based at least in part on one or more selection criteria, at a dynamically adjusted level of aggressiveness of the selection performance. The scrubber logic controller may adjust the level of aggressiveness of the selection performance, based at least in part on a determined time left to complete the selection performance at a current level of aggressiveness. Other embodiments may be described and/or claimed.