Hi,
This question still pop up from customers about the right ratio. For some case the customer would prefer to have 1:1 relation between vCPU and pCPU. MY understand it can depend on CPU clock cycle more than the number of cores. Will the highs clock cycle will expose less ready time and CPU contention ? Am I assuming right ?
Thnak you
This topic comes up from time to time. Below are some related discussions that may help shed some light on the matter:
How to calculate vCPU:pCPU ratio
Often times VDI desktop pools are a common ground where ratios can be higher as the idle processing across these deployments often times allow for a higher ratio compared to a cluster that would have some production workloads like exchange, database or application workloads that require CPU demand.
It depends on the workload more than anything. If you have a bunch of servers that mainly run at 25% cpu utilization with high cpu spikes here and there, you can run at a higher ratio then servers that run at 75% CPU utilization all the time. CPU ready is the time the virtual wanted to schedule something but couldn't.
This topic comes up from time to time. Below are some related discussions that may help shed some light on the matter:
How to calculate vCPU:pCPU ratio
Often times VDI desktop pools are a common ground where ratios can be higher as the idle processing across these deployments often times allow for a higher ratio compared to a cluster that would have some production workloads like exchange, database or application workloads that require CPU demand.
Thank you for the helpful answer, When it commes to CPU ready time what is the factor that will cause the relation between CPU and vCPU schudel , Is is HZ or cores regardless what is the speed of the CPU ?