Priority Task Offloading Decision Optimization Based on Multi-Objective Grey Wolf Algorithm in Internet of Things Edge Computing Scenarios

In Internet of Things (IoT) application scenarios, the traditional method of prioritizing task offloading based on scalar information such as maximum tolerable latency fails to meet the processing requirements of urgent tasks. To ensure the prioritized handling of urgent tasks, this paper proposes a priority task offloading decision optimization strategy based on the multi-objective Grey Wolf Algorithm. The research focuses on the problem of priority task offloading decision and considers the edge server task processing program cache. The optimization objectives, including minimizing overall latency, social loss rate, and load imbalance, are established to formulate a multi-objective optimization task offloading decision model. The multi-objective Grey Wolf Algorithm is employed to solve the problem. Experimental results demonstrate that the proposed multi-objective Grey Wolf Algorithm effectively reduces overall latency and social loss rate, optimizes load balancing among edge servers, and outperforms other algorithms significantly.