Research on Delay-aware Scheduling Algorithms for Edge Task Migration in High-concurrency Environments

Edge computing systems often suffer from elevated task response delays under high-concurrency conditions. Conventional scheduling approaches struggle to accommodate both task urgency and the dynamic nature of node resources, frequently resulting in delayed migration decisions and system bottlenecks. To address these challenges, this paper proposes a delay-aware task migration scheduling algorithm. This approach proposes a multi-dimensional latency evaluation function. By incorporating both real-time task urgency and node load status, it dynamically assesses the potential gains of task migration and determines the optimal execution timing. Simulation results demonstrate the algorithm’s advantages in reducing average response time, minimizing ineffective migrations, and improving scheduling stability, making it suitable for heterogeneous, high-density edge computing scenarios.