A Hybrid Strategy Integrating Ant Colony Optimization and Locust-Inspired Algorithm (HACO-LA) to Boost Efficiency and Performance in cloud Resource Management

Efficient cloud resource management is vital, as it ensures the accurate selection and allocation of resources to diverse workloads or applications. This entails real-time balancing of workload performance, compliance, and cost to achieve optimization. Since there are many cloud users involved in scheduling tasks, or cloudlets, the scheduling process becomes complex. It involves selecting appropriate data centers, servers (hosts), and virtual machines (VMs). There are several bioinspired algorithms that are effective and popular for solving the NP-complete problem of cloudlet scheduling. By using these algorithms, cloudlets can be efficiently allocated to achieve faster execution times, better resource utilization, and a shorter waiting time. This study presents a hybrid technique that combines the strengths of both ant colony optimization and locust-inspired algorithm (HACO-LA), which have been shown to outperform other bio-inspired algorithms in cloud computing environments. Implementing this technique will lead to a decrease in the average response time, while also increasing the utilization of VMs and servers. The method was evaluated using the Cloud Sim toolbox with authentic data. It was compared to the preexisting Artificial Bee Colony Optimization (ABC) algorithm and Particle Swarm Optimization (PSO) algorithm. The findings demonstrated that HACO-LA surpassed both methods, leading to notable enhancements in server utilization, increased reliability, and decreased average response time.