A Study on Multi-Objective Cloud-Edge Collaborative Task Scheduling Method for Edge Networks

As the expansion of new services continues to grow, there may be a shortage of equipment computing, storage and communication resources. At the same time, tasks have the possibility of computation and transmission failure when device resources are limited. Therefore, this paper proposes a research on a multi-objective cloud-edge cooperative task scheduling method for edge networks. The method fully considers the multi-objective optimisation requirements such as task execution efficiency, delay, resource utilisation and energy consumption, and designs an adaptive scheduling strategy by constructing a dynamic model of tasks and resources. The strategy combines genetic algorithms and deep reinforcement learning techniques to achieve reasonable distribution and collaborative processing of tasks between cloud centres and edge nodes. On the basis of the service configuration mechanism, a processing model for microservices is established. By analysing the constraints in task scheduling, priority constraints, capacity constraints and execution time constraints are pointed out. And the scheduling scheme is selected for the task by a multi-criteria decision-making method based on fuzzy logic. The experimental results show that the proposed method has significant advantages in improving the task execution efficiency, reducing the delay and optimising the resource utilisation, which effectively meets the demand of multi-objective optimisation in edge networks, and provides new ideas and references for task scheduling in edge computing environments.