The CPU Wait Time Per Dispatch alarm is raised when the average queue time for the virtual machine waiting for CPU to become available exceeds a threshold.
The average queue time should remain under 60,000ns. This alarm is raised when the average queue time exceeds 60 microseconds. A high severity alarm is raised when the average queue time exceeds 100 microseconds.
When the alarm is raised there is CPU starvation.
- For Windows Server see the Windows Server | CPU Drilldown | Virtualized CPU page | CPU wait time per dispatch chart.
- See your Hyper-V administrator for ways to address this issue.