Application of predator-prey optimization for task scheduling in cloud computing

Cloud computing environments require scheduling to allocate resources efficiently and ensure optimal performance. It is possible to maximize resource utilization and minimize execution time by scheduling cloud systems effectively. Meta-heuristic algorithms aim to address this NP-hard problem by taking into account these QoS parameters. In order to deal with the task scheduling problem, we utilize a new meta-heuristic algorithm known as Predator-Prey Optimization (PPO). In PPO, predators and preys are modeled and their energy gains are determined by their body mass and interactions. Faster convergence rates enhance PPO’s ability to find optimal solutions. The balance between exploration and exploitation makes it suitable for solving real-world problems in unknown spaces. The PPO-based Task Scheduling algorithm (PPOTS) has the goal of reducing execution time and makespan while increasing resource utilization. In this study, the PPOTS algorithm is compared to five well-known meta-heuristic algorithms: Whale Optimization Algorithm (WOA), Salp Swarm Algorithm (SSA), Spotted Hyena Optimization Algorithm (SHO), Grasshopper Optimization Algorithm (GOA), and Sooty Tern Optimization Algorithm (STOA). Furthermore, the proposed PPOTS algorithm was compared with two new meta-heuristic based scheduling algorithms, and showed a better performance than the other two algorithms. Resource utilization and execution cost are enhanced by 8% and 15%, respectively, through the proposed method.