Hybrid Optimization algorithm with the combination of PSO and genetic algorithm for task scheduling in cloud computing

Abstract

Cloud computing is an aligned distributive architecture consisting of a network of associated, digitized machines that are dynamically provisioned as individual computing resources depending on SLAs (service-level agreements) between customers and vendor of services. It provides easy, on-demand access to shared hardware, applications, and data pools. This technology allows businesses and individuals to store and analyze data in third-party data centers. While various load balancing techniques for cloud computing have been introduced recently, they do not always achieve the intended outcomes. It offers simple, usage-based accessibility to shared hardware instruments apps, and data pools. Businesses and individuals can store and analyze data in independent data centers thanks to this technology. Numerous techniques to balance load in cloud settings have been presented recently, but they don’t always produce the intended outcomes. This work presents a hybrid optimization solution for load balancing that integrates Particle Swarm Optimization (PSO) with genetic algorithms. The efficiency level of the presented solution is assessed according to consumed energy, duration of response, and the overall migrations, comparing it to the ACO and PSO after implementation in MATLAB. The findings indicate that the new solution surpasses existent optimization solutions for every evaluated criterion.