A Dynamic Task Scheduling Approach based on Arithmetic Optimization Algorithm in Cloud System

One of the most important challenges in cloud computing is task scheduling. Virtual Machines (VMs) differ in terms of configuration and platform due to heterogeneous underlying platforms, task interdependencies, and task interdependencies. Meta-heuristic approaches are preferred for NP-complete problems such as task scheduling. This paper proposes a method to solve task scheduling problems using an Arithmetic Optimization Algorithm called AOATS by considering both cost computation minimization and makespan minimization. The efficiency of the proposed AOATS is shown by comparisons with some existing meta-heuristic algorithms such as Gannet Optimization Algorithm (GOA), Grey Wolf Algorithm (GWO), and Cat Swarm Algorithm (CSA). Performance comparisons of the novel heuristic scheduling algorithm against other existing algorithms demonstrated its superiority on metrics such as makespan, degree imbalance, and cost.

Keywords:

Task scheduling , Cloud computing , Makespan , Computation cost