A Multi-Strategy Improved Tuna Swarm Optimization Algorithm for Cloud Task Scheduling

Metaheuristic algorithms are often used to optimize the allocation and utilization of cloud computing resources for users' tasks. And a good task scheduling strategy is crucial for the efficiency and QoS of cloud computing services. This paper proposes a multi-strategy improved tuna swarm optimization (MITSO) algorithm for the cloud task scheduling, which takes both makespan and load balancing into account of optimization. A series of strategies such as chaotic maps, levy flight, and opposition-based learning are taken to address the limitations of the original TSO algorithm, such as the premature convergence. The effectiveness of the proposed MITSO algorithm was verified using the CloudSim simulation platform compared with other metaheuristic algorithms such as PSO, WOA and GWO. Experiment results show that the proposed MITSO has a better global optimization capability.