Abstract The rapid growth of Cloud computing is accompanied by a significant increase in the consumed energy by data centers. This huge increase in energy consumption has become a major concern because of both its costs and environment impact. Consolidating virtual machines (VMs) in minimum numbers of physical machines (PMs) is an effective way to reduce energy consumption. Various algorithms and techniques have been proposed to improve energy efficiency via VM consolidation. However, recent findings of Green Peace showed that energy demands of data centers are continuing to grow annually. This continuous increase queries the area of reducing energy consumption of data centers. As an attempt to figure out the reasons behind the gap between Green Peace recent findings on energy consumption and the proposed VM consolidation techniques and to bridge this gap, we undertake this survey. We discuss these techniques by highlighting their strengths and limitations and we justify the need for taking into consideration the energy overhead created by both the VM migrations and PMs transitions in and out of the low power mode.