Service Migration Strategy Considering Task Deadline in Edge Computing
Qiushi Cao · 2023
In mobile edge computing, due to the mobility of connected autonomous vehicles (CAV), service migration is required when the vehicle travels to a new location. Making the best migration decision will be a great challenge due to the uncertainty of the network environment, deadline constraints, and resource constraints. In this paper, we aim to optimize the service migration strategy for CAV users with different deadline tasks. We try to design a suitable service migration strategy to maximize the number of completed tasks under the constraints of service migration energy consumption and user offloading costs based on the deadlines of different types of tasks. Next, we propose an improved algorithm for particle swarm optimization that can better search for service migration strategy in late iteration. We use the proposed mobility model to simulate the generation of movement trajectories. Extensive experimental results demonstrate that our proposed solution can perform more tasks than other algorithms and can provide higher quality of service to CAV users.