The invention discloses a hierarchical
edge computing unloading method based on priority. The method comprises a
resource allocation and task scheduling strategy taking time
delay minimization as an optimization target. On one hand, a layered
edge computing framework is constructed for an emergency scene, edge nodes which are closer to
user equipment and have fewer computing resources process simple tasks so as to quickly return warning information, and servers which are farther from the
user equipment and have more computing resources process complex tasks so as to finally return accurate results. And on the other hand, a
resource allocation and task scheduling optimization scheme based on the service emergency priority is provided for the problem that the time
delay is increased due to limited resource competition, so that the total time
delay of the
system is minimized. A dynamic priority task scheduling
algorithm (DPTSA) on a
server is designed in order to ensure that the time delay of high priority tasks is still minimum under the condition that the tasks are very dense. Therefore, the user can obtain real-
time response as soon as possible in case of emergency.