Since traffic diversity and volume increase with growing popularity of mobile applications, there is the strong need to manage the traffic carried by networks. Software defined networks can simplify network management while enabling new services by employing traffic management including routing whose goal is to maximize the given utility while satisfying capacity requirements. Another key concept to meet up huge data traffic is cloud-based radio access networks. By integrating cloud services to radio access networks, operators will make use of network functions virtualization which allows to host different virtualized functions on a common hardware platform. In this thesis, an efficient routing algorithm is proposed to minimize the cost based on power consumption determined by the number of active OpenFlow switches and active links in a software defined networks while satisfying throughput requirements of all flows according to constraints on link capacities in the software defined mobile network. The algorithm is also implemented in mobile network by combining resource allocation in a cloud radio access network. The performance of the proposed algorithm is evaluated based on power consumption efficiency for different network topologies with various scenarios.