You are on page 1of 1

Use as few vCPUs as possible.

This provides a potential performance improvement for both the


virtual machine and the overall host.
Unused vCPUs still consume timer interrupts.
Maintaining a consistent memory view among multiple
vCPUs consumes resources.
Some older guest operating systems execute idle loops on
unused vCPUs, thereby consuming resources that might
otherwise be available for other uses.
The guest scheduler might migrate a single-threaded
workload among multiple vCPUs, thereby losing cache
locality.

You might also like