International audienceEnergy savings are among the most important topics concerning Cloud and HPC infrastructuresnowadays. Servers consume a large amount of energy, even when their computingpower is not fully utilized. These static costs represent quite a concern, mostlybecause many datacenter managers are over-provisioning their infrastructures comparedto the actual needs. This results in a high part of wasted power consumption. In thispaper, we proposed the BML (“Big, Medium, Little”) infrastructure, composed of heterogeneousarchitectures, and a scheduling framework dealing with energy proportionality.We introduce heterogeneous power processors inside datacenters as a way to reduceenergy consumption when processing variable workloads. Our framework brings an intelligentutilization of the infrastructure by dynamically executing applications on thearchitecture that suits their needs, while minimizing energy consumption. In this paperwe focus on distributed stateless web servers scenario and we analyze the energy savingsachieved through energy proportionality