Cache Page Allocation Using Hit Ratio Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing method of evenly allocating page resources in a memory pool reduces the hit ratio of cache objects with large resource requirements, leading to a decreased overall memory hit ratio and increased service processing delay.
Innovation Solution
A resource allocation method that uses a processor to create models based on page resources and data read hit ratios to dynamically adjust resource allocation, considering factors like reuse distance, memory eviction information, and historical data, to optimize resource distribution among cache objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If page resources are evenly allocated to multiple cache objects based on remaining memory pool resources, then allocation simplicity is maintained, but the hit ratio of cache objects with large resource requirements is reduced
Solution Approach 1:
The patent applies local quality by differentiating resource allocation strategies for different cache objects based on their individual characteristics. Instead of uniform allocation, the system evaluates each cache object's resource requirements and allocates page resources proportionally, ensuring that cache objects with larger needs receive more resources to maintain their hit ratios.
Solution Approach 2:
The patent implements dynamic resource allocation where the memory management unit continuously monitors cache object performance and adjusts page resource distribution in real-time. When memory pool resources change, the system dynamically recalculates and redistributes page resources to maintain optimal hit ratios for all cache objects, particularly protecting those with large resource requirements.
2Device complexity
If data is evicted from cache objects in the same proportion when memory pool resources are insufficient, then resource reclamation is simplified, but the overall memory hit ratio is reduced
Solution Approach 1:
The patent applies local quality to eviction operations by treating different cache objects differently based on their resource requirements and performance characteristics. Instead of uniform proportional eviction, the system selectively evicts data from cache objects that can tolerate resource reduction while protecting cache objects that are critical for maintaining overall hit ratio.
Solution Approach 2:
The patent implements feedback mechanisms where the memory management unit monitors cache hit ratios and uses this information to guide eviction decisions. When resources need to be reclaimed, the system feedback-drivenly adjusts which cache objects undergo eviction and to what extent, ensuring that overall hit ratio is maintained while still achieving necessary resource reclamation.
3Device complexity
If uniform resource allocation is used, then system complexity is reduced, but service processing delay increases
Solution Approach 1:
The patent implements dynamic resource allocation that adapts to changing system conditions and cache object requirements. By continuously monitoring performance metrics and adjusting page resource distribution in real-time, the system optimizes service processing speed without requiring complex manual configuration, achieving both low delay and manageable complexity through automated adaptation.
Data Source
AI summary
A resource allocation method includes a processor that outputs a first model based on a page resource allocated to each cache object and a data read hit ratio of the cache object, where the page resource allocated to each cache object is from a memory pool provided by a storage system. The first model indicates a page resource to be allocated to the cache object when a specific hit ratio is met. The processor outputs a second model based on the first model corresponding to each cache object, where the second model indicates a target page resource to be allocated to each cache object. The processor adjusts, based on the second model, the page resource allocated to each cache object. The processor adjusts the page resource of the cache object based on the second model.