Combined approach for task scheduling in cloud computing systems using meta-heuristic algorithms and virtual machine migration

Purpose This study aims to optimize task scheduling systems in cloud computing environments by leveraging efficient meta-heuristic algorithms to maximize hardware efficiency, minimize space utilization, and reduce maintenance costs. Cloud computing has emerged as a new information technology platform beyond traditional information systems that can meet as many customer requirements as possible. The advantages of cloud computing for task scheduling systems include reduced use of hardware, efficient use of data center space and low maintenance costs. However, meta-heuristic algorithms are more efficient. Design/methodology/approach In this paper, a combined approach based on the virtual machine migration (VM migration) algorithm and two popular meta-heuristic algorithms, namely, genetic algorithm (GA) and particle swarm optimization (PSO), is proposed to simultaneously take advantage of both heuristic and meta-heuristic techniques for obtaining a balanced exploration–exploitation search process in the allocation of virtual machines (VMs) to servers in cloud computing task scheduling systems. On the one hand, VM migration is a heuristic method with a local search strategy (exploitation) that is used to balance the load and minimize the energy consumption of the system. On the other hand, the applied meta-heuristics are population-based algorithms with powerful global search mechanisms (exploration). Findings The results show the higher efficiency of the proposed method compared to other traditional methods. By performing the proposed algorithm, not only the energy consumption of the entire system is reduced compared to other methods due to the shutdown of idle servers based on the migration function, but also carbon production from cooling devices and greenhouse gas emissions are effectively reduced. Originality/value The main unique contribution of this research is the development of a hybrid heuristic-meta-heuristic algorithm that effectively integrates VM migration with GA and PSO, achieving a balanced exploitation–exploration strategy for optimal VM allocation. This approach enhances resource utilization and energy efficiency in cloud computing systems while ensuring high-quality service delivery. The proposed approach indirectly contributes to carbon emission reduction by lowering IT energy consumption, which, when combined with standard power usage effectiveness and emission factors, translates into reduced environmental impact.