Cloud computing has officially entered the commercial application stage, which puts forward higher requirements on network load balancing. Leveraging effective load distribution and traffic scheduling algorithm to reasonably allocate the request data between every processing nods to achieve optimal processing capacity of the system is one of the effective ways to improve the utilization of network resources. The unique self-directed learning and reconfiguration capabilities of cognitive network [1] enable the load balancing to become more effective. Based on research of the existing traffic scheduling algorithm, this paper improves the weighted least connections scheduling algorithm, and designs the Adaptive Scheduling Algorithm Based on Minimum Traffic (ASAMT). ASAMT conducts the real-time minimum load scheduling to the node service requests and configures the available idle resources in advance to ensure the service QoS requirements. Being adopted for simulation of the traffic scheduling algorithm, OPNET is applied to the cloud computing architecture. Experimental results show that, under the premise of no large network cost, the load condition of this algorithm is better than that of the unmodified weighted least connection scheduling algorithm.