Cloud computing has indeed revolutionized the IT industry by eliminating the need for businesses to invest in physical hardware, relying instead on companies that offer cloud services. However, the energy-intensive nature of cloud data centers, heavily reliant on electricity, poses significant challenges, especially with recent spikes in electricity prices. To address this issue, a novel approach involving Extreme Gradient Boosting (XGBoost) is proposed. The XGBoost model is leveraged to optimize data placement and node scheduling, effectively offloading or moving storage within data centers. Additionally, it predicts electricity prices, a crucial factor in data center operation costs, enabling efficient resource allocation. This project's performance evaluation relies on a real-world dataset from the Independent Electricity System Operator (IESO) in Ontario, Canada. By leveraging machine learning techniques, the aim is to reduce energy consumption, enhance data center efficiency, and ultimately lower operational costs. With data split into 70% for training and 30% for testing, the XGBoost model holds promise in mitigating the energy challenges faced by modern data centers, contributing to a more sustainable and cost-effective cloud computing infrastructure.

KEYWORDS: Data Storage, Energy Saving, Electricity Price Forecasting, XGBoost.