CPU resources are handled differently in Hyper-V than they are VMware ESXi and XenServer. In VMware and XenServer, if a vCPU slows down, all the other vCPUs will have to slow down, because they are all scheduled or time-sliced together. Microsoft addresses the scheduling problem inside of the guest or virtual machine. As the virtual machine is aware that it is running in a guest within a hypervisor, the operating system knows to schedule processes independently as opposed to combining with vCPU scheduling. Hyper-V understands CPU calls will be coming independently. Oversubscription is not as big of a deal as this method has much less overhead.
A vCPU in Hyper-V is simply a time slice of the physical CPU resource. You can overcommit vCPU resources in Hyper-V as would be expected in any hypervisor. Independent vCPU management is handled automatically by Hyper-V. You can reserve vCPU processing time, assign vCPU processors, and assign weights to vCPU processing priorities...