Optimizing Cloud Computing Load Balancing Through Extended Ant Colony Optimization

Due to the rapid advancement of IT and the widespread adoption of Internet applications, there is a significant and increasing interest among companies and developers in cloud computing. Among the crucial challenges in cloud computing, virtual machine allocation stands out as a prominent issue that has garnered significant attention. However, earlier studies tended to overlook the vital aspect of resource load balancing. In this article, our objective is to address the load balancing problem across all physical machines and virtual machines within the cloud computing, with the aim of maximizing resource utilization. In order to accomplish this, we developed an effective virtual machine (VM) allocation approach employing ant colony optimization in response to the problem’s NP-hardness. The ant colony optimization strategy is expanded in our method to particularly address the virtual machine allocation situation. The fundamental ant colony optimization method has also been augmented using a physical machine selection strategy. This strategy aims to avoid premature convergence and escape from local optima, leading to more effective and robust VM allocation results. This paper presents a novel cloud load balancing policy inspired by Extended Ant Colony Optimization (EACO), drawing inspiration from Ant Systems. The primary objective is to minimize the makespan and energy utilization, resource utilization, and virtual machine allocation associated with load balancing. The study involves the simulation of the EACO algorithm using the CloudSim toolkit, and the outcomes demonstrate the superior performance of cloud load balancing when utilizing the EACO approach.