The customer is charged for the pre-allocation of resources to an organization virtual datacenter. A provider may charge the customer based upon the allocated resources, the guaranteed resources, or a combination of the guaranteed resources and the consumption of resources beyond the guarantee. The cloud provider manages resource overcommitments by defining a guaranteed percentage of allocated resources, such as CPU, memory resources, virtual CPU speed, and an optional, maximum virtual machine limit. Only the vCloud provider can expand or contract resources. The Allocation Pool model may consume resources from one or more resource pools, depending on the vCloud provider’s configuration.

The Allocation Pool model enables an organization to procure resources under normal operating conditions but has the capability to burst for more resources when need arises. Because this model guarantees a specified percentage of the allocated resource, the remainder that is not guaranteed is shared with other tenants. This can result in resource contention.

The Allocation Pool model is usually a good fit for relatively steady state workloads that occasionally surge.