Communication-aware, energy-efficient VM placement in cloud data center using ant colony optimization

The many advantages of the cloud computing trend are being recognised by an increasing number of tech-savvy companies and industry leaders utilising this technology to run their businesses, provide better customer service, and significantly boost their overall profit margins. In a cloud computing environment, users submit their jobs to run on virtual machines instead of a cloud host. Mapping virtual machines to physical machines is known as virtual machine placement (VMP). There may be communication among the VMs, which may lead to network congestion. Hence, VMP has significant importance with respect to resource utilization, energy consumption and total communication cost in data centres. In this work, we have applied multi-objective ant colony optimization for the VMP problem to minimize communication cost, energy usage and resource wastage. The proposed work is simulated, and analysis is done on how these factors affect each other and vary with the increasing load. Numerous experiments are performed with varying resource correlation coefficients in the requested VMs for performance analysis of this work. As observed, the proposed ant colony optimization based metaheuristic algorithm outperforms multi-objective genetic algorithm, other ant colony optimization based, well-known heuristic-based first fit decreasing algorithms, and random-based approaches in terms of overall power consumption, resource wastage and the total network data transfer rate. © 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management.