Performance on a new private cloud deployment is not meeting expectations. An engineer has determined the virtualization host is experiencing high ready/wait times. VMs with the following…

Decreasing the CPU count for a VM might reduce its share but could degrade its performance if it is truly compute-bound, and it doesn't directly address the overall contention unless the VM was grossly over-provisioned.

Increasing the memory size for VM2 could potentially reduce its disk I/O or other resource bottlenecks, allowing it to complete tasks faster and free up CPU resources sooner. This indirect effect can reduce overall CPU ready/wait times for the host by decreasing the time VMs spend actively consuming CPU while waiting for other resources.

Decreasing the memory size for VM3 would likely increase paging or swapping, putting more strain on storage I/O and potentially increasing its CPU usage, exacerbating high ready/wait times.

Increasing the CPU count for VM4 would add more demand for CPU resources from the host, directly worsening CPU contention and leading to even higher ready/wait times across the host.