Analysis of SMP VM CPU Scheduling

System virtualization allows resources of a single phys-ical machine to be shared among multiple virtual ma-chines (VMs). This can increase system flexibility, but it can also invalidate some assumption that the OS run-ning in a VM has about the system it is executing on. This paper surveys the implementation of symmetric multiprocessing (SMP) in a virtualized environment. It examines two proposals for optimizing the implemen-tation of SMP in VMs, and it analyzes the implemen-tation of SMP VMs for two popular virtualization sys-tems: VMware ESX server and Xen. The benchmark results show that SMP VMs do experience greater over-head than uniprocessor VMs for computationally inten-sive workloads. However, optimizations in ESX server and Xen reduce the overhead below what it would be for the most naive implementation. 1