The fifth generation mobile network(5G) is coming,and the base station network carries more and more pressuredue to the higher rate demand of users. Load balancing trafficis crucial for high link utilization and low latency. Most of theresearch on load balancing algorithms focuses on data center sce-narios and works based on TCP. Traditional algorithm used in 5G base station network is Equal Cost Multipath Routing(ECMP),which has poor performance due to its congestion agnostic nature. Other existing cogestion-aware algorithms work well in datacenter networks. However, the transport protocol and the traffictype is different in 5G base station network. It will change theperformance of the original algorithms and cause the challengesto implement the schemes. In this paper, we propose RavenFlow,an effective load balancing algorithm working on layer 2 whileusing UDP. It makes use of different congestion information andeliminates the dependence on reverse packets. We evaluate the performance of RavenFlow in ns-3 simulator under typical 5Gtraffic. The result shows that RavenFlow achieves shorter Flow Completion Time(FCT) and smaller difference of utilization fordifferent links.