A Hybrid Sand Cat Swarm Optimization and Particle Swarm Optimization (SCSO–PSO) Algorithm for DDoS-Resilient Task Scheduling in Cloud Computing Environments

Abstract

Cloud computing has become a cornerstone of modern computational infrastructure, enabling scalable, on-demand resource provisioning for diverse applications. However, the efficiency and reliability of cloud services are increasingly threatened by Distributed Denial-of-Service (DDoS) attacks, which severely degrade system performance by overloading network and computational resources. Task scheduling, recognized as an NP-hard optimization problem, becomes particularly challenging under such adversarial conditions, where conventional metaheuristic algorithms often suffer from premature convergence and stagnation in local optima. To address these limitations, this study proposes a novel hybrid algorithm that integrates Sand Cat Swarm Optimization (SCSO) with Particle Swarm Optimization (PSO) referred to as SCSO–PSO for DDoS-resilient task scheduling in cloud computing environments. The proposed algorithm exploits SCSO's diversified exploration capability, inspired by the low-frequency hunting behavior of sand cats, while leveraging PSO's rapid convergence dynamics through a stagnation-driven switching mechanism. The framework is implemented within the CloudSim simulation toolkit and evaluated under both normal and DDoS-affected scenarios, using makespan, resource utilization, and execution time as primary performance metrics. Comparative analysis against benchmark algorithms including Genetic Algorithm (GA), PSO, Artificial Bee Colony (ABC), and the recent GA–PSO hybrid demonstrates that SCSO–PSO consistently achieves lower makespan and improved scheduling stability under adversarial conditions. The findings contribute a robust scheduling framework that enhances the resilience of cloud computing systems against evolving Cybersecurity threats.