Machine Learning-Based Cache Replacement Policies: A Survey

Despite extensive developments in improving cache hit rates, designing an optimal replacement policy that mimics Belady’s algorithm still remains a challenging task. Existing standard static policies does not adapt to the dynamic nature of memory access patterns, and diversity computer programs only exacerbates problem. Several factors affect design such as hardware upgrades, overheads, model latency, etc. The amalgamation fundamental concept like with advanced machine learning algorithms provides surprising results drives development towards cost-effective solutions. In this paper, we review some machine-learning based outperformed heuristics.