The distribution network extending from the grid to the social service domain needs to support more services. As applications continue to increase with new business expansion, this may result in insufficient resources for edge devices. In order to support the scheduling and collaborative processing of tasks given a resource constraint from the edge device level, a cloud-edge collaborative online scheduling strategy for distribution station area tasks with a microservice architecture is proposed. This paper analyzes the task scheduling problem in the cloud-edge containerized scenario from the task and device perspectives, designs the priority and task assignment strategy based on the cloud-edge containerization scheduling mechanism of power tasks, and schedules the tasks in real time using an improved online algorithm. Simulation results show that the proposed strategy has higher task execution efficiency and can improve the completion rate of important tasks and system security when the resources of edge devices are limited through resource swapping. This work is supported by the National Key Research and Development Program of China (No. 2016YFB0900600).