Software-Defined Networking(SDN)adapts logically-centralized control by decoupling control plane from data plane and provides the efficient use of network resourc-es.However,due to the limitation of tradition-al routing strategies relying on manual config-uration,SDN may suffer from link congestion and inefficient bandwidth allocation among flows,which could degrade network perfor-mance significantly.In this paper,we propose EARS,an intelligence-driven experiential net-work architecture for automatic routing.EARS adapts deep reinforcement leing(DRL)to simulate the human methods of leing expe-riential knowledge,employs the closed-loop network control mechanism incorporating with network monitoring technologies to realize the interaction with network environment.The proposed EARS can le to make better con-trol decision from its own experience by inter-acting with network environment and optimize the network intelligently by adjusting services and resources offered based on network re-quirements and environmental conditions.Under the network architecture,we design the network utility function with throughput and delay awareness,differentiate flows based on their size characteristics,and design a DDPG-based automatic routing algorithm as DRL de-cision brain to find the near-optimal paths for mice and elephant flows.To validate the net-work architecture,we implement it on a real network environment.Extensive simulation results show that EARS significantly improve the network throughput and reduces the aver-age packet delay in comparison with baseline schemes(e.g.OSPF,ECMP).