VMware Engine node types

A Google Cloud VMware Engine private cloud is a collection of vSphere clusters created with a set of dedicated isolated bare metal hardware nodes in a region. These nodes provide the compute, memory, and storage required to run VMware ESXi, and are the basic unit of consumption of VMware Engine. This document details the node types, their resource capacity and availability in the regions.

A vSphere cluster must have nodes with identical storage. Google Cloud VMware Engine allows creation of a cluster with a mix of hyperconverged infrastructure (HCI) and storage only node types. You can create a cluster with ve2-standard-128 and ve2-standard-so node types, but not mix two different HCI node types.

Node types

The following sections list node types that are available when creating a VMware Engine private clouds or clusters.

Hyperconverged node types

The following table lists HCI node types that are available when creating a VMware Engine private cloud or a cluster.

Node type vCPUs/Node Memory/Node (GiB) Storage/Node (TB)*
ve1-standard-72 72 768 19.2
ve2-standard-96† 96 2048 25.5
ve2-standard-128† 128 2048 25.5
ve2-mega-64† 64 2048 51.2
ve2-mega-96† 96 2048 51.2
ve2-mega-128† 128 2048 51.2

Storage-only node types

The following table lists storage-only node types that are available when creating a VMware Engine private cloud or cluster.

Node type Storage/Node (TB)*
ve1-standard-so 19.2
ve2-standard-so† 25.5
ve2-mega-so† 51.2

* Raw storage excluding cache.
† Available in select regions.

Region availability

The following table lists the regional availability of Google Cloud VMware Engine node types.

Node types Regions
ve1-standard-72
ve1-standard-so
  • asia-northeast1
  • asia-south1
  • asia-south2
  • asia-southeast1
  • australia-southeast1
  • europe-west2
  • europe-west3
  • europe-west4
  • europe-west6
  • europe-west8
  • europe-west12
  • me-west1
  • northamerica-northeast1
  • northamerica-northeast2
  • southamerica-east1
  • southamerica-west1
  • us-east4
  • us-central1
  • us-west2
  • us-south1
ve2-standard-96
ve2-standard-128
ve2-mega-96
ve2-mega-128
ve2-standard-so
ve2-mega-so
  • australia-southeast1
  • us-east4
  • us-central1
  • europe-west3
  • southamerica-east1
  • southamerica-west1
ve2-mega-64
  • australia-southeast1

Node identification

Each VMware Engine private cloud has a nodeTypeId that's available as soon as you create a VMware Engine private cloud. Retrieve the nodeTypeId with the projects.locations.nodeTypes.get REST request.

Node billing

There are two options for paying for VMware Engine nodes: On-demand and commitment. When you provision a VMware Engine node, you control the VMs on that host. For more information about pricing, see VMware Engine pricing.

Node commitments

You can purchase VMware Engine nodes with a one- or three-year committed use discount (CUD) where you make a fixed monthly payment, or you can pay for nodes as you use them. Once you purchase a CUD, you can't cancel it. For more information about pricing, see VMware Engine pricing.

Purchasing nodes as CAPEX or OPEX

Google does not provide guidance about accounting. Review the details about the VMware Engine node product and seek guidance from your Controllership team and auditors before you make any accounting decisions regarding CAPEX or OPEX.

Node maintenance

If a node fails, VMware Engine immediately and automatically adds a new node to the affected VMware cluster to restore service operability. VMware Engine also sends an email to the primary address notifying you about the replacement of the failed node and the name of the private cloud. For more information about maintenance, see Private cloud maintenance and updates.

What's next