HWOA: an intelligent hybrid whale optimization algorithm for multi-objective task selection strategy in edge cloud computing system

Edge computing is a popular computing modality that works by placing computing resources as close as possible to the sensor data to relieve the burden of network bandwidth and data centers in cloud computing. However, as the volume of data and the scale of tasks processed by edge terminals continue to increase, the problem of how to optimize task selection based on execution time with limited computing resources becomes a pressing one. To this end, a hybrid whale optimization algorithm (HWOA) is proposed for multi-objective edge computing task selection. In addition to the execution time of the task, economic profits are also considered to optimize task selection. Specifically, a fuzzy function is designed to address the uncertainty of task's economic profits and execution time. Five interactive constraints among tasks are presented and formulated to improve the performance of task selection. Furthermore, some improved strategies are designed to solve the problem that the whale optimization algorithm (WOA) is subject to local optima entrapment. Finally, an extensive experimental assessment of synthetic datasets is implemented to evaluate the multi-objective optimization performance. Compared with the traditional WOA, the diversity metric (A-spread), the hypervolume (HV) and other evaluation metrics are significantly improved. The experiment results also indicate the proposed approach achieves remarkable performance compared with other competitive methods.