CMS Computing experience at LHC

After the first years of LHC data taking, from cosmic runs to proton-proton and heavy ion physics at 7 and 8 Te V, the Computing Model of the CMS experiment was finally verified under real operating conditions. The Model had been prepared and tested since 2005, on top of Grid-compliant computing centers and services, under the overall service and site coordination offered by the WLCG Collaboration. With the growth in maturity of the Grid services and the evolution of the computing technologies available, the CMS Computing Model started a natural evolution over time that encountered a major boost with the massive experience on data processing, archiving, data transfer and distributed data analysis coming from LHC operations. In synergy with this process, the WLCG Collaboration launched Technical Evolution Groups (TEG) to assess the situation and develop a medium-long term evolution strategy in the adopted technologies to support the LHC experiments use-cases at the best. In this contribution we will describe the core of the original CMS Computing Model, its major achievements during the LHC operation, and its proposed evolution in the next years.