MUSTACHE: Multi-Step-Ahead Predictions for Cache Eviction

TL;DR

MUSTACHE is a novel cache replacement algorithm that uses deep learning to predict future page requests, leading to improved cache hit ratios and reduced read/write operations compared to traditional heuristics.

Contribution

It introduces a deep learning-based multi-step-ahead prediction approach for cache eviction, integrating forecasting into cache management for the first time.

Findings

Outperforms exact LRU in cache hit ratio by 1.9%.

Reduces cache misses' read/write operations by over 10%.

Demonstrates effectiveness on benchmark datasets.

Abstract

In this work, we propose MUSTACHE, a new page cache replacement algorithm whose logic is learned from observed memory access requests rather than fixed like existing policies. We formulate the page request prediction problem as a categorical time series forecasting task. Then, our method queries the learned page request forecaster to obtain the next $k$ predicted page memory references to better approximate the optimal B\'el\'ady's replacement algorithm. We implement several forecasting techniques using advanced deep learning architectures and integrate the best-performing one into an existing open-source cache simulator. Experiments run on benchmark datasets show that MUSTACHE outperforms the best page replacement heuristic (i.e., exact LRU), improving the cache hit ratio by 1.9% and reducing the number of reads/writes required to handle cache misses by 18.4% and 10.3%.