Abstract—Cloud computing seems to be currently the hottest new trend in data storage, processing, visualization, and analysis. There has also been a significant rise in cloud computing as government organizations and commercial businesses have migrated toward the cloud system. It has to do with dynamic resource allocation on demand to provide guaranteed services to clients. Another of the fastest-growing segments of computer business involves cloud computing. It was a brand-new approach to delivering IT services through the Internet. This paradigm allows consumers to access computing resources as in puddles over the Internet. It is necessary and challenging to deal with the allocation of resources and planning in cloud computing. The Random Forest (RF) and the Genetic Algorithm (GA) are used in a hybrid strategy for virtual machine allocation in this work. This is a supervised machine-learning technique. Power consumption will be minimized while resources are better distributed and utilized, and the project’s goal is to maximize resource usage. There is an approach that can be used to produce training data that can be used to train a random forest. Planet Lab’s real-time workload traces are utilized to test the method. The suggested GA-RF model outperformed in terms of data center and host resource utilization, energy consumption, and execution time. Resource utilization, Power consumption, and execution time were employed as performance measures in this work. Random Forest provides better results compared with the Genetic Algorithm.
 
Keywords—cloud computing, infrastructure-as-a-service, random forest, genetic algorithm, service level agreements, optimization, resource allocation


Copyright © 2023 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.