Improving Data Center Efficiency: A Combinatorial Optimization Framework for Resource Allocation

This work addresses resource allocation optimization in data centers, formulated as a variation of the bin packing problem. Due to the high costs of on premises infrastructures, strategies for migrating workloads to the public cloud and shutting down physical and virtual machines are investigated to enhance operational cost efficiency. Preliminary experiments demonstrate the effectiveness of the proposed algorithm at different scales, highlighting its performance and limitations in complex scenarios. Execution time metrics and the number of simulations indicate that the algorithm is efficient for small to medium-sized tasks but faces challenges under extensive demands.