We are going to propose a new approach to implement predictive Web page ranking in cloud environment. User query is transferred to the server side of cloud based search engine. Searching and ranking procedures have been executed on the server side and final output is sent to the client as a response. In this paper, prediction task has been performed based on user behaviour. Available data, resources, and functional support systems have been processed using schedulers. Session, duration, and other user relevant data are combined with the processed information for prediction. Different Web based data repositories are considered as cloud structure for energy stabilization.