vSphere Metrics Watch: Three Key Processor Metrics
The compute tier is responsible for executing operations and supports running the code that provides an environment for the virtual machines. The compute layer is made up of servers, each two one or more processors and with each of those processors bearing one or more processing cores. Note: For an overview of memory monitoring see the first article in this series.
CPU usage provides an administrator with an at-a-glance look at how far the virtual environment is being pushed while the CPU ready metric is a measure for how long it takes for enough physical processor resources to be freed up to service a virtual machine’s needs. As this article is focused on overarching performance metrics, I decided to focus on the most obvious performance-related metric.
With that said, what is good and what is bad with regard to CPU usage in a virtual environment? Many agree that a consistent physical CPU utilization metric that stays well above 75% – 80% is the point at which an administrator should begin to consider adding processor resources. Even though that still leaves 20% of room, bear in mind that many loads still need room to “peak” at certain times, so it’s important to leave a little room.
Bear in mind that many loads still need room to “peak” at certain times
Your vSphere environment might have plenty of available processor compute capacity, but you may still run into a resource contention issue known as CPU ready. CPU Ready is the percentage of time that a virtual machine is ready to perform a task, but is unable to schedule enough physical CPU resources to actually get the work done. If the CPU Ready percentage gets too high, noticeable latency can be the result. In general CPU contention the approaches 10% or higher means that you may start to experience significant problems, but the exact threshold is dependent on the kinds of workloads you’re running.
Co-Stop is related to CPU Ready, but at a somewhat different level. Co-Stop deals with virtual machines that have multiple vCPUs assigned. If you Co-Stop value goes above 10%, your virtual machine is experiencing contention when attempting to schedule resources to satisfy all vCPUs. The more vCPUs you add to a host that is already reaching CPU capacity, the more likely it is that you will experience this issue in a multi-vCPU environment.
If you don’t actually need multiple vCPUs in a virtual machine, don’t add them.
Work has to get done. That’s the job of the CPU in a virtual machine. Understanding key performance issues around CPU resourcing is critical to ensuring that your workloads operate as required by the business.