CacheBoost: Harnessing Machine Learning for Peak Cache Performance
Sharath Kumar Jagannathan, Maheswari Raja, P. Vijaya, Reena Abraham · Advances in computer science research · 2024
This research investigates the integration of machine learning (ML) models into cache management systems to enhance overall performance.Two distinct strategies, the Block Cache model and Vector Cache model, are implemented, each incorporating widely used cache replacement policies-Least Recently Used (LRU) and Least Frequently Used (LFU).Furthermore, three ML models-Logistic Regression, K-Nearest Neighbors (KNN), and Neural Network-are integrated into these cache systems.The primary goal is to improve the cache hit rate by combining ML models with Belady's Optimal algorithm.The performance of the five cache models is assessed using key metrics such as cache hit rate, miss rate, and eviction rate.A comparative analysis is undertaken to gauge the effectiveness of each approach and the influence of various ML models on cache performance.This study aims to provide valuable insights into the complex interaction between traditional cache replacement policies and advanced ML techniques, offering a nuanced understanding of the potential enhancements in cache hit rates achieved through machine learning integration.The findings and observations contribute to the ongoing exploration of cache optimization, guiding future developments to enhance system performance.