When you create a virtual machine (VM) on Compute Engine, you specify a machine series and a machine type for the VM. Each machine series is associated with one or more CPU platforms. If there are multiple CPU platforms available for a machine type, you can select a minimum CPU platform for the VM.
A CPU platform offers multiple physical processors, and each of these processors are referred to as a core. For all processors available on Compute Engine, a single CPU core can run as multiple hardware multithreads through Simultaneous multithreading (SMT), which is known on Intel processors as Intel Hyper-Threading Technology. On Compute Engine, each hardware multithread is called a virtual CPU (vCPU). When vCPUs are reported to the VM as occupying different virtual cores, Compute Engine ensures that these vCPUs never share the same physical core.
The machine type of your VM specifies its number of vCPUs, and you can infer its number of physical CPU cores using the default vCPU per core ratio for that machine series:
- For the Tau T2D and Tau T2A machine series, VMs always have one vCPU per core.
- For all other machine series, VMs have two vCPUs per core by default.
- You can optionally set a VM to have one vCPU per core instead of two vCPUs per core, which might benefit some workloads. Importantly, when you do this, the machine type of your VM no longer reflects the correct number of vCPUs. Instead, the pricing and number of physical CPU cores remains the same as it would be for the default two vCPUs per core ratio, and the number of vCPUs is half of value indicated by the machine type.
Arm processors
For Arm processors, Compute Engine uses one thread per core. Each vCPU maps to a physical core with no SMT.
The following table describes the Arm processors that are available for Compute Engine VMs.
| CPU processor | Processor SKU | Supported machine series | All-core sustained frequency (GHz) |
|---|---|---|---|
| Ampere Altra | Q64-30 | 3.0 |
x86 processors
For most x86 processors, each vCPU is implemented as a single hardware thread. The Tau T2D machine series is the exception, with one vCPU representing one physical core.
Intel processors
On Intel Xeon processors, Intel Hyper-Threading Technology supports multiple threads running concurrently on each core. The specific size and shape of your VM instance determines the number of its vCPUs.
| CPU processor | Processor SKU | Supported machine series | Base frequency (GHz) | All-core turbo frequency (GHz) | Single-core max turbo frequency (GHz) |
|---|---|---|---|---|---|
| Intel Xeon Scalable Processor (Ice Lake) 3rd Generation |
Intel® Xeon® Platinum 8373C Processor | 2.6 | 3.4 | 3.5 | |
| 2.6 | 3.4 | 3.5 | |||
| Intel Xeon Scalable Processor (Cascade Lake) 2nd Generation |
|||||
| Intel® Xeon® Gold 6268CL Processor | 2.8 | 3.4 | 3.9 | ||
| Intel® Xeon® Gold 6253CL Processor | 3.1 | 3.8 | 3.9 | ||
| Intel® Xeon® Platinum 8280L Processor | 2.5 | 3.4 | 4.0 | ||
| Intel® Xeon® Platinum 8273CL Processor | 2.2 | 2.9 | 3.7 | ||
| Intel Xeon Scalable Processor (Skylake) 1st Generation |
Intel® Xeon® Scalable Platinum 8173M Processor | 2.0 | 2.7 | 3.5 | |
| Intel Xeon E7 (Broadwell E7) | Intel® Xeon® E7-8880V4 Processor | 2.2 | 2.6 | 3.3 | |
| Intel Xeon E5 v4 (Broadwell E5) | Intel® Xeon® E5-2696V4 Processor | 2.2 | 2.8 | 3.7 | |
| Intel Xeon E5 v3 (Haswell) | Intel® Xeon® E5-2696V3 Processor | 2.3 | 2.8 | 3.8 | |
| Intel Xeon E5 v2 (Ivy Bridge) | Intel® Xeon® E5-2696V2 Processor | 2.5 | 3.1 | 3.5 | |
| Intel Xeon E5 (Sandy Bridge) | Intel® Xeon® E5-2689 Processor | 2.6 | 3.2 | 3.6 |
*N2 machine types that have 80 or more vCPUs use the Intel Ice Lake CPU.
AMD processors
AMD processors provide optimized performance and scalability using SMT. In almost all cases, Compute Engine uses two threads per core, and each vCPU is one thread. Tau T2D is the exception where Compute Engine uses one thread per core and each vCPU maps to a physical core. The specific size and shape of your VM instance determines the number of its vCPUs.
| CPU processor | Processor SKU | Supported machine series | Base frequency (GHz) | Effective frequency (GHz) | Max boost frequency (GHz) |
|---|---|---|---|---|---|
| AMD EPYC Milan 3rd Generation |
AMD EPYC 7B13 | 2.45 | 2.8 | 3.5 | |
| AMD EPYC Rome 2nd Generation |
AMD EPYC 7B12 | 2.25 | 2.7 | 3.3 |
Frequency behavior
The previous tables describe the hardware specifications of the CPUs that are available with Compute Engine, but keep the following points in mind:
- Frequency: A PC's frequency, or clock speed, measures the number of cycles the CPU executes per second, measured in GHz (gigahertz). Generally, higher frequencies indicate better performance. However, different CPU designs handle instructions differently, so an older CPU with a higher clock speed can be outperformed by a newer CPU with a lower clock speed because the newer architecture deals with instructions more efficiently. For more information about CPU clock cycles and performance, see Clock rates and system performance,
- Base frequency: The frequency at which the CPU runs when the system is
idle or under light load. When running at its base frequency, the CPU draws
less power and produces less heat.
- A VM's guest environment reflects the base frequency, regardless of what frequency the VM is actually running at.
- All-core turbo frequency: The frequency at which each CPU typically
runs when all cores in the socket are not idle at the same time. Different
workloads place different demands on a system's CPU. Boost technologies
address this difference and help processes adapt to the workload demands by
increasing the CPU's frequency.
- Most VMs get the all-core turbo frequency, even if only the base frequency is advertised to the guest environment.
- Ampere Altra Arm processors can provide more predictable performance because the frequency for Arm processors is always the all-core turbo frequency.
- Max turbo frequency: The frequency a CPU targets when stressed by a demanding application like a video game or design modeling application. It's the maximum single-core frequency a CPU achieves without overclocking.
- Processor power management technologies: Intel processors support multiple
technologies to optimize the power consumption. These technologies are divided
into two categories, or states:
- C-states are states when the CPU has reduced or turned off selected functions. C-state is only supported for C2 machine types.
- P-states provide a way to scale the frequency and voltage at which the processor runs so as to reduce the power consumption of the CPU. For VMs other than C2, C-State and P-State are not supported currently, so idling virtual CPUs within the guest environment might not work as expected.
What's next
- Learn more about Machine families.
- Learn more about Virtual machine instances.
- Learn more about Images.
- Learn how to Specify a minimum CPU platform.
Try it for yourself
If you're new to Google Cloud, create an account to evaluate how Compute Engine performs in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.
Try Compute Engine free