Which statement accurately describes compute resource planning for a host computer?

Which statement accurately describes compute resource planning for a host computer?

• A. The host does not need to have enough compute resources to support VM workload.
• B. The host must have enough compute resources to support the VM workload.
• C. The host must run a supported operating system.
• D. The number of Windows VMs must be known.

Answer: (B)

The key idea here is that compute resource planning centers on ensuring the host has enough CPU, memory, and I/O capacity to run the virtual machines that will be placed on it, including the overhead of the hypervisor. You size the host based on the total resources your VMs will demand, plus some headroom for peak loads and failover, so there’s no contention or severe performance degradation. If the host doesn’t have sufficient resources, VM performance will suffer and service levels can be breached. The other points aren’t the primary focus of compute resource planning. Having a supported operating system is important for compatibility, but it’s not about resource sizing. Knowing the exact number of Windows VMs at planning time isn’t strictly necessary either, since workloads can scale_up or scale_out and you plan around expected demand and capacity, not a fixed count.

The key idea here is that compute resource planning centers on ensuring the host has enough CPU, memory, and I/O capacity to run the virtual machines that will be placed on it, including the overhead of the hypervisor. You size the host based on the total resources your VMs will demand, plus some headroom for peak loads and failover, so there’s no contention or severe performance degradation. If the host doesn’t have sufficient resources, VM performance will suffer and service levels can be breached.

The other points aren’t the primary focus of compute resource planning. Having a supported operating system is important for compatibility, but it’s not about resource sizing. Knowing the exact number of Windows VMs at planning time isn’t strictly necessary either, since workloads can scale_up or scale_out and you plan around expected demand and capacity, not a fixed count.