Optimization of Energy Consumption in Data Centers Using Deep Learning: A Systematic Literature Review

This paper presents a systematic literature review (SLR) on the use of deep learning techniques to reduce energy consumption in data centers. The main objectives are to identify key metrics, optimization strategies, and the main challenges addressed in the literature. The Scopus database was used to retrieve relevant articles, and a structured protocol was applied to ensure replicability and reliability. The review highlights three main research axes: energy-efficient workload allocation, cooling system optimization, and predictive maintenance. It also reveals gaps in experimental validation, standardization of evaluation metrics, and integration with real-time systems. Future research should focus on developing interpretable models, establishing regulatory frameworks, and testing solutions in real-world environments.