Blocking Cache Line Selection With Dual Binary Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cache replacement algorithms in blocking cache systems often select recently replaced cache lines, neglecting those that have not been accessed for a long time, thereby affecting performance.
Innovation Solution
A method and device that utilize a blocking state binary tree in conjunction with a usage state binary tree to select the earliest replaceable cache line, ensuring accurate identification of cache lines to be replaced.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional cache replacement algorithms are used in blocking cache systems, then the implementation is simple, but the cache hit rate decreases because recently replaced cache lines are selected instead of those not accessed for a long time
Solution Approach 1:
The cache replacement algorithm is segmented into two independent binary trees: a usage state binary tree that tracks access patterns and a blocking state binary tree that manages replacement eligibility. This segmentation allows the system to independently track usage history and replacement status, enabling accurate selection of cache lines that have not been accessed for a long time while maintaining implementable complexity through modular tree structures.
Solution Approach 2:
The usage state binary tree acts as an intermediary data structure that mediates between cache line access events and replacement decisions. By introducing this intermediate tracking mechanism, the system can identify cache lines based on their usage history without requiring complex direct analysis of access patterns, thus improving hit rate while keeping the replacement logic manageable.
2Measurement precision
If the cache system tracks detailed usage history to select the earliest replaceable cache line, then the accuracy of replacement selection improves, but the hardware resource consumption increases
Solution Approach 1:
The algorithm extracts only the essential replacement eligibility information into a separate blocking state binary tree, pulling out the critical replacement tracking function from the general cache management logic. This extraction allows precise identification of replaceable cache lines using dedicated blocking state bits without requiring the entire cache system to maintain complex usage histories for all lines, thereby reducing overall hardware resource consumption while maintaining high selection accuracy.
Solution Approach 2:
The solution adds a temporal dimension to cache line tracking by maintaining usage state information that records when cache lines were last accessed. By introducing this time-based dimension through the usage state binary tree, the system can accurately identify the earliest replaceable cache line based on usage history without needing to track every detail of cache line behavior, achieving precise replacement selection with manageable hardware overhead.
Data Source
AI summary
The present disclosure relates to the computer technology and provides a method and a device for selecting a cache line to be replaced, and a storage medium. In the present disclosure, a blocking state binary tree is added to replace a replacement state binary tree, and a cache line to be replaced is selected based on the blocking state binary tree and a usage state binary tree, such that the earliest cache line, acting as the cache line to be replaced, can be accurately selected from the replaceable cache lines in a blocking cache.