Task scheduling directly affects the performance of cloud computing systems because due to underutilisation and overutilisation of resources, wastage of resources and degradation in performances have been witnessed, respectively. Therefore, metaheuristics have been commonly employed to optimise scheduling and improve task execution efficiency. However, most of the existing approaches struggle to attain optimal or near-optimal performance due to the complexity and dynamic nature of the tasks. To efficiently handle this, the present work explores the employability of the Walrus Optimization Algorithm (WaOA) for task scheduling due to its simplicity, adaptability, and efficient exploration–exploitation capability. The simulated results reveal the overwhelming capability of WaOA for task scheduling relative to other metaheuristics in terms of convergence speed, accuracy, and resource utilisation. These results advocate the employability of WaOA for task scheduling.