The recent Conferences of Parties (COP) contributed to launch initiatives to reduce the global warming bound to greenhouse gases. In the IT world, data centers are one of the major energy consumers. This consumption is due to the various components of the dbms (database management system). query optimizer are the most energy consumers, because they are traditionally designed to improve the performance of requests. The existing works integrate energy into the design of the queries optimizers by considering the sequential mode to process operations. Recently, dbms centralized propose the parallel mode involving several processors (or core) to optimize the response time of requests, but neglect the energy dimension. In this article, we propose an analytical model cost that estimates the energy consumption of queries executed in parallel mode. This model is integrated into the queries processing module of PostgreSQL to predict the energy consumption.