MILP Model for Energy- Efficient Scheduling in Cloud Data Center

In the Era of Industry 4.0, the shift from physical infrastructure to cloud data centers has become a growing trend among companies and enterprises of various scales. However, the growth of modern cloud data centers has raised concerns about the energy consumption of these systems. Several existing studies on energy-efficient scheduling have proposed linear programming models before introducing heuristic algorithms to solve the problem. These models are typically straightforward and sometimes do not fully represent the mathematical aspects of the problem's constraints. Our research aims to provide a deeper mathematical insight into the energy-efficient scheduling problem. This paper focuses on heterogeneous cloud data centers employing virtualization technology and predefined job requests, including release time, deadline, and execution time. We propose a Mixed Integer Linear Programming (MILP) model to optimize the scheduling solution. The model includes an objective function built to compute the optimal solution, along with constraints represented by equations and inequalities to ensure the solution meets the requirements of problem. Additionally, Dynamic Power Management (DPM) is integrated into the model to further reduce the energy consumption. Experimental results demonstrate that integrating the proposed MILP model with the DPM technique significantly reduces energy consumption compared to the model that do not apply DPM.