AI assisted multi-objective cloud computing model for optimized task scheduling and enhanced QOS

Abstract

In recent decades, cloud computing has gained popularity due to the extensive collection of autonomous systems with a flexible framework and diversified features. Various communities require a cloud computing paradigm to maximize their revenue due to its commercial reality. Scheduling of resources to the cloud consumers dramatically influences the cost-benefit of the service providers. Several kinds of research have already been made, focusing on task scheduling and resource utilization. Job shop scheduling is one of the strong NP-hard problem for the production of optimal scheduling strategies. Evolutionary algorithms such as genetic algorithm and tabu search have been emerged to perform optimal job scheduling in cloud computing environments, but that are confined to perform a single objective. Hence to meet the multiple objectives in cloud computing platforms, we proposed a novel artificial intelligence-based task scheduling strategy to facilitate minimum makespan, energy efficiency, reduced computational cost, and reliability. The proposed modified sheep flock heredity algorithm (MSFHA) facilitates the optimal task scheduling strategy by selecting the job schedules with the Longest job to the High-speed processor (LJHP), Smallest job to the High-speed processor (SJHP), and high-affinity values.