vcpus

vCPUs, or virtual CPUs, are the CPU resources presented to a virtual machine by a hypervisor. A vCPU represents a virtual execution context that the guest operating system can schedule tasks on. In most virtualization environments, each vCPU maps to a thread of a physical processor or to a virtualized execution unit, and multiple vCPUs can be assigned to a VM to enable parallel processing.

Hypervisors such as VMware ESXi, KVM, Hyper-V, and Xen schedule vCPUs on the host’s physical CPUs. The mapping is influenced by oversubscription, CPU quotas, and scheduling policies, as well as contention for memory bandwidth and cache. In practice, 1 vCPU often corresponds to a single hardware thread on a modern CPU, but definitions vary by vendor and platform. Some providers define vCPU more loosely, treating a vCPU as a fraction of a core or as a different allocation unit altogether. Consequently, a given number of vCPUs in one environment may not translate directly to the same number of physical cores on another.

In cloud computing and virtualization, vCPUs are a primary dimension of instance flavor or SKU, alongside memory and I/O capacity. Users specify the number of vCPUs when provisioning virtual machines or containers, and licensing for certain software may be tied to vCPU counts. Performance depends on factors such as host load, NUMA topology, memory bandwidth, and CPU scheduling, so capacity planning and monitoring of metrics like CPU ready time and throttling are important for ensuring acceptable guest performance.