Current researches for Internet of Things (IoT) QoS mainly focuses on the formulation of service level protocols, which improves some performance of resource scheduling, but there are still many shortcomings in resolving the real-time and personalized requirements of IoT. Aiming at the hierarchical resource scheduling algorithm of the IoT, the key issues of hierarchical resource scheduling are analyzed in detail. The hierarchical resource scheduling of the Internet of Things based on improved heuristic algorithm is deeply studied and explored. A cuckoo search algorithm based on adaptive Cauchy mutation is proposed. Because the algorithm is prone to premature, easy to fall into the local optimal solution, and unable to find the global optimal solution, by introducing mutation operator, the improved algorithm has a certain ability of local random search, while accelerating convergence to the optimal solution in the later period, maintaining the diversity of solutions. The simulation results show that the average service success rate of the proposed resource scheduling algorithm is close to 99%, which can effectively guarantee the relative fairness of user requests, meet the real-time and personalized needs of different users, and improve the utilization rate of resources.