Using Processing Features for Allocation of Virtual Machines in Cloud Computing

Resource scheduling is a key process for a cloud computing platform, which generally uses virtual machines (VMs) as scheduling units. The use of virtualization techniques provides great flexibility with the ability to instantiate multiple VMs on one physical machine (PM), migrate them between the PMs and dynamically scale VM's resources. In this work we proposed an architecture to handle the allocation of VMs based on the processing power for heterogeneous clouds, where there is a wide variety of CPU types. We proceeded a case study with extensive replications of eight experiments using a real private cloud. The results show that it is possible to use the proposed approach to define a processing unit, supported by the CPU usage limitation, to enable the VM's processing power remain at the same level regardless the PM.