This research presents a novel approach to address the Virtual Machine Placement Problem (VMPP) in cloud data centers with the aim of minimizing energy consumption. The main contributions of this study are threefold. Firstly, a Duelist Algorithm specifically designed for VMPP, which introduces a unique concept of duelists combined with optimization techniques. The algorithm aims to strike a balance between exploration and exploitation in the search space, leading to more effective resource allocation and energy-efficient cloud data center management. Secondly, enhance the performance of the Duelist Algorithm by reducing the number of algorithm-specific parameters. This simplifies the implementation process and increases the algorithm's adaptability to various real-world problems, making it more user-friendly and robust. Lastly, conduct a comprehensive comparison of the Duelist Algorithm with the widely used Hybrid Harmony Search Algorithm (HS+SA+LS) in terms of energy consumption and overall efficiency. The experimental results demonstrate that the Duelist Algorithm consistently outperforms the Hybrid Harmony Search Algorithm, achieving remarkable improvements in both best and mean fitness values. Additionally, the Duelist Algorithm exhibits lower standard deviation values, indicating more stable and consistent performance. The findings of this research validate the effectiveness of the proposed Duelist Algorithm in minimizing energy consumption and optimizing cloud resource allocation. The reduction of algorithm-specific parameters further contributes to its versatility and simplicity.

Data Center energy usage has risen dramatically because of the rapid growth and demand for cloud computing. This excessive energy usage is a challenge from an economic and environmental point. Virtual Machine Placement (VMP) along with virtualization technologies is widely used to manage power utilization in data centers. The assignment of virtual machines to physical machines affects energy consumption. VMP is a process of mapping VMs onto a set of PMs in a data center to minimize total power consumption and maximize resource utilization. The VMP is an NP-hard problem due to its constraints and huge combinations. In this paper, we formulated the problem as a single objective optimization problem in which the objective is to minimize the energy consumption in cloud data centers. The main contribution of this paper is hybrid and adaptive harmony search algorithm for optimal placements of VMs to PMs. HSA with adaptive PAR settings, simulated annealing and local search strategy aims at minimizing energy consumption in cloud data centers with satisfying given constraints. Experiments are conducted to validate the performance of these variations. Results show that these hybrid HSA variations produce better results than basic HSA and adaptive HSA. Hybrid HS with simulated annealing, and local search strategy gives better results than other variants for 80 percent datasets.