Cloud computing is the evolutionary step to transform a large part of the Information and Communication Technology industry. It is the result of the efforts to provide the opportunity to focus on hardware and software cost and the impending reduction of preservation. Cloud computing constitutes both business and an economic model which has been gaining popularity awareness in industry. Cloud computing services providers hope that the widespread adoption of the cloud will bring them more profit and they are actively promoting the technology. Energy Cost of the data-centers in the Cloud computing efficiency one of the main area which needs to be focus. In this paper we summarize previous offerings to the discussion of energy cost of cloud computing, provide a working definition of energy cost of cloud computing and discuss its importance. In this research work, we present an adaptive energy cost saving framework in the cloud to achieve and maintain best Service Level Agreements (SLA). We propose lightweight approach to accurately estimate the power usage of virtual machines and cloud servers at each geographical location. We consider both the system power usage and the SLA requirements, and influence the learning techniques to achieve optimal resource allocation and optimal power efficiency.