Performance bounded energy efficient virtual machine allocation in the global cloud

•We analyze the effects of global virtual machine allocation on energy consumption, using a variety of real-world policies and a realistic testing scenario.•We found that by using an allocation policy designed to minimize energy, total energy consumption could be reduced by up to 14%, and total monetary energy costs could be reduced by up to 26%.•We have begun performance qualification of our energy cost driven allocation policies through network capability tests.•Our results indicate that performance and IaaS provider implementation costs have a significant influence on selection of optimal virtual machine allocation policies.