A Data Center (DC) can be defined as a cyber- physical “complex” system since within it computing and energy aspects coexist together with their interdependencies. Due to this dual aspect, the measurement of DC energy efficiency is an articulated and open issue, because it involves both Information Technology (IT) and Cooling Technology perspectives. Therefore, the analysis of energy efficiency in DCs, through a set of globally accepted metrics, is an ongoing challenge. In particular, the area of productivity metrics – that are parameters referred to the actual processes of computing, storage and data transfer – needs to be further explored. Moreover, till now none of the existing proposed metrics provides a direct measure of the useful work in a DC. To this purpose, this work aims to investigate and analyze the relationship between different types of server workloads and power consumed by the relative computing nodes (core) used. The paper aims to make a step forward beyond state of the art in productivity metrics and also to server performance and power management, by analyzing the server energy consumption behavior through statistical data analysis. To this purpose in this work real data obtained by experimental campaigns on the ENEA-C.R. Portici facilities have been analyzed.