Traditional network management policy built on the theory of random network model has some disadvantages such as redundant management data

high probability of synchronized data collecting and so on.The shortcomings will lead management performance degradation.So a self-adaptive policy based on scale-free network is mentioned to avoid disadvantages of traditional network management.The policy dynamically probes the rank of the entity based on the threshold of the metric.It can adjust automatically the period of the data collectors according to the rank of the network entity.Experiments confirm that the self-adaptive policy can reduce amount of collected data and eliminate the synchronization of data collecting.The policy improves management performance and keeps the same function of traditional network management.