Evaluation of Resource Allocation in Cloud using Machine Learning

Abstract

Cloud computing has revolutionized the way computing the resources are allocated and managed, and it offers scalability, flexibility, and the cost savings. Proper resource allocation, however, remains a difficult problem due to varying workloads, unpredictable demand, and the need for optimal performance. Machine learning (ML) techniques have been recognized as a promising solution to optimizing the resource allocation by predicting workload patterns, optimizing resource utilization, and reducing latency. In this paper, we have compared the various ML-based framework for the resource allocation in the cloud computing environments on the basis of their efficiency in order to improve the efficiency and the cost control. Through comparative evaluation, we highlight the merits and demerits of different ML models, including the contextually of their suitability in the actual implementations. The results reveal that the proposed model achieves accuracy (Decision Tree 100%, AdaBoost 72.2%, Support vector machine 98.5%, logistic regression 97.6% and Gradient boosting 100%).