Due to the advent of Internet of Things(IoT) plethora of services has been emerged and to perpetuate all these services using cloud computing paradigm is really tiresome. A new promising service provider called as Fog computing came into the picture where the distance between Iot and edge device is small to provide the services efficiently to the end users. There are different considerable factors like service response time, and expected QoS must met without violating other resource constraints. In this paper we try to layout an architecture which is based on the combination of cloud and fog by introducing a middleware called as cloud fog control middleware, which manages the service request according to some constraints. By using this architecture we can maximize the advantages of next generation computer system, however the architecture require new strategies to manage the mapping of services to resources. Despite several Heuristic and Meta Heuristic techniques has been proposed by different authors to solve the service placement in fog computing with considering different parameters like quality of service(QoS), Latency, etc. In this paper we are trying to minimize the energy consumption in fog computing paradigm by formulating the service placement plan in order to utilize the resources efficiently by considering the Active and Idle state of machine. First we calculate the energy consumption by the application(number of tasks) requesting for a particular service. Undoubtedly the service placement problem is a combinatorial optimization problem. The optimal solution obtained distribute the load to some other fog node by appropriate placement of service which leads to reduce the over heat generated by a particular fog node. In this way we can achieve energy minimization in fog computing.