A vCPU, or virtual CPU, is a unit of processing capacity that a cloud provider assigns to a virtual machine or database instance. It represents a share of a physical processor that the hypervisor schedules for your workload. Cloud services commonly size and price instances by how many vCPUs they include.
What a vCPU equals in hardware varies by provider and instance type. On some platforms it corresponds to a hardware thread, and on others to a full core, and processor generation and speed differ too. Two vCPUs from different providers, or even different instance families, may not deliver the same performance.
Because of this, the number of vCPUs is a starting point for sizing, not a precise measure of speed. Real performance depends on the workload, memory, storage and network, and on whether the instance shares resources with others.
To size a service, check how the provider defines a vCPU for your instance type, start with an estimate based on your workload, and measure actual usage. Adjust up or down depending on observed CPU use, and compare cost per unit of work rather than only the vCPU count.