Multi-port Cache LRU Ranking via Threshold Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-port cache memories face challenges in implementing Least Recently Used (LRU) policies when multiple cache accesses occur in a single access cycle, leading to potential evictions of recently used cache entries due to delayed promotions, which can increase power consumption and die size with complex circuitry.
Innovation Solution
A cache controller circuit that maintains rank values and a threshold value to identify and promote the most at-risk cache way in a single access cycle, reducing the risk of eviction without requiring complex logic circuits by comparing rank values to the threshold and modifying them accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If complex circuitry is used to track and promote cache ways in multi-port access, then cache management accuracy is improved, but device complexity and power consumption increase
Solution Approach 1:
The cache is divided into multiple independent cache ways, each with its own rank value. The ranking system is segmented into discrete levels (first rank, second rank, etc.) that can be independently tracked and updated without requiring complex inter-dependent logic circuits.
Solution Approach 2:
Rank values are pre-calculated and maintained for each cache way based on access patterns. When a cache hit occurs, the promotion decision is made immediately using these pre-established rank values, eliminating the need for complex real-time analysis circuits.
2Reliability
If complex promotion logic is implemented to handle multiple cache hits in one access cycle, then cache way promotion accuracy is improved, but power consumption increases
Solution Approach 1:
Instead of implementing full complex logic to handle all possible promotion scenarios, the system uses a simplified approach: compare rank values to a threshold and promote if the condition is met. This partial action approach achieves sufficient promotion accuracy without the power cost of exhaustive logic.
Solution Approach 2:
The system changes the parameter used for promotion decisions from complex temporal analysis to simple rank value comparison. By maintaining rank values as a separate parameter and comparing them against a threshold, the system achieves accurate promotion with minimal power consumption.
3Loss of time
If traditional LRU tracking is used in multi-port cache, then eviction decisions are delayed, but cache way promotion timeliness improves
Solution Approach 1:
Rank values are maintained continuously and updated with each access, so when a cache hit occurs, the promotion decision can be made immediately using the already-calculated rank values, eliminating delays in promotion timing.
Solution Approach 2:
The traditional mechanical LRU tracking system that requires complex circuitry to determine least recently used cache ways is replaced with a numerical ranking system. Cache ways are assigned rank values that directly indicate their eviction priority, allowing rapid eviction decisions without complex mechanical tracking logic.
Data Source
AI summary
An apparatus includes a cache controller circuit and a multi-ported cache memory including a plurality of cache ways. The cache controller circuit is configured to maintain rank values and a threshold value usable to classify the rank values. A given rank value corresponds to a least recently used one of the plurality of cache ways. The cache controller circuit is further configured to receive, in a common access cycle, first and second memory access requests for the cache memory, and, in response to a determination that the first and second memory access requests correspond to respective first and a second cache ways, compare the corresponding rank values for the first and second cache ways to the threshold value. The cache controller circuit is further configured to, based on the comparison, modify the rank value of a selected one of the first and second cache ways.