Modernizing cloud computing systems with integrating machine learning for multi-objective optimization in terms of planning and security

Abstract

Cloud enterprises face challenges in managing large amounts of data and resources due to the fast expansion of the cloud computing atmosphere, serving a wide range of customers, from individuals to large corporations. Poor resource management reduces the efficiency of cloud computing. This research proposes an integrated resource allocation security with effective task planning in cloud computing utilizing a Machine Learning (ML) approach to address these issues. The suggested ML-based Multi-Objective Optimization Technique (ML-MOOT) is outlined below: An enhanced task planning, based on the optimization method, aims to reduce make-span time and increase throughput. An ML-based optimization is developed for optimal resource allocation considering various design limitations such as capacity and resource demand. A lightweight authentication system is suggested for encrypting data to enhance data storage safety. The proposed ML-MOOT approach is tested using a separate simulation setting and compared with state-of-the-art techniques to demonstrate its usefulness. The findings indicate that the ML-MOOT approach outperforms the present regarding resource use, energy utilization, reaction time, and other factors.