6.2. Overcommitting Virtualized CPUs

The KVM hypervisor supports overcommitting virtualized CPUs. Virtualized CPUs can be overcommitted as far as load limits of guest virtual machines allow. Use caution when overcommitting VCPUs as loads near 100% may cause dropped requests or unusable response times.
Virtualized CPUs (VCPUs) are overcommitted best when a single host physical machine has multiple guest virtual machines, where the guests do not share the same VCPU. The Linux scheduler is very efficient with this type of load. KVM should safely support guest virtual machines with loads under 100% at a ratio of five VCPUs (on 5 virtual machines) to one physical CPU on one single host physical machine. KVM will switch between all of the machines making sure that the load is balanced.
You cannot overcommit symmetric multiprocessing guest virtual machines on more than the physical number of processing cores. For example a guest virtual machine with four VCPUs should not be run on a host physical machine with a dual core processor. Overcommitting symmetric multiprocessing guest virtual machines in over the physical number of processing cores will cause significant performance degradation.
Assigning guest virtual machines VCPUs up to the number of physical cores is appropriate and works as expected. For example, running guest virtual machines with four VCPUs on a quad core host. Guest virtual machines with less than 100% loads should function effectively in this setup.

Important

Do not overcommit memory or CPUs in a production environment without extensive testing. Applications which use 100% of memory or processing resources may become unstable in overcommitted environments. Test before deploying.
For more information on how to get the best performance out of your virtual machine, refer to the Red Hat Enterprise Linux 6 Virtualization Tuning and Optimization Guide.