VCF as a Service pricing

IBM Cloud® for VMware Cloud Foundation as a Service compute resources are available as either a single-tenant or multitenant model.

The VMware Cloud Foundation (VCF) as a Service single-tenant model offers dedicated hosts that are managed by your organization. First, you order a VCF as a Service single-tenant Cloud Director site then you create one or more virtual data centers (VDCs) where you deploy your virtual machines (VMs).

For the multitenant model, IBM® hosts and operates a resource pool where you order virtual resources as needed. IBM manages your capacity, performance, and hardware upgrades. When you order a multitenant instance, you select the location to create a VDC in an existing Cloud Director site.

On the VMware Cloud Foundation as a Service order page, click See pricing details to view the rate card that outlines cost details by feature.

Understanding billing terminology

Usage

Metering determines how IBM counts usage for the full potential size of the resource for the time period that the resource is used.

For example, a single-zone on-demand VDC is created with a resource limit of 100 virtual CPU (vCPU) and 800 GB RAM. The data center has no VMs running on it, so you do not receive a charge for the vCPU and RAM. If an 8 vCPU with 8 GB virtual machine (VM) is started, metering is calculated for the size of that VM. If the VM uses fewer resources than the ones assigned to it, metering is applicable to the full size of the VM.

Allocation

IBM bills based on the resources that you allocate when you provision an instance. The allocated resources are guaranteed.

Metering is applicable to the full potential size of the resource for the life of the resource.

For example, a single-zone reserved VDC is created with a resource allocation of 100 vCPU and 800 GB RAM and no VMs are created or running on it. Metering is applicable to 100 vCPU and 800 GB RAM.

Peak metric usage

The metric displays on your bill as either monthly peak metric usage or hourly peak metric usage.

Monthly peak usage is the maximum value of the metric that is used over a full month. For example, a single-zone reserved VDC is created with 100 vCPU and 800 GB RAM. Later in the month, the data center is reduced to 50 vCPU and 400 GB RAM. The monthly peak usage is 100 vCPU and 800 GB RAM.

Hourly peak usage is the maximum value of the metric that is used over an hour. For example, if 100 vCPU is used for a minute of the hour with 0 vCPU used for the other 59 mins, the hourly peak metric usage is 100 vCPU.

View service billing and usage details in the VMware Solutions console. Click Manage > Billing and usage to review the service usage and charges. In the Billing and usage view, locate the VMware Cloud Foundation as a Service service type. Then, locate the Organization plan to find the service usage across all VDCs in that organization. For multitenant models, VDC usage is located in a separate plan for either the on-demand or the reserved pricing plan.

Billing considerations for single-tenant models

Review minimum order requirements and storage considerations for single-tenant models.

Minimum 2 servers (minimum 7 servers for vSAN™ profiles)

2 hosts (2-socket 32 cores, 192 GB RAM)
24 TB of shared storage
Minimum of 1 VDC with a Medium Dedicated Edge

With a seven host minimum with vSAN profiles, a high availability standard with FTT=2 and RAID6 ensures that a system can tolerate multiple failures during a maintenance event with no data loss. Consider the overhead that you must incorporate when you determine the amount of storage to provision.

Storage that is ordered after the first day of the month is prorated for the first month. Storage that is removed during the month is charged for the full month that it is removed. Each storage change is treated individually. For example, when you remove storage then add storage, the bill reflects both storages for that month. For vSAN host profiles, the profile name displays the total of both raw and the usable storage.

Single-tenant instance billing
Feature	Billing model
Host profiles	Size for bare metal server profiles including sockets, core, and RAM. Hosts are billed monthly. Hosts that you order after the first day of the month are prorated for the first month. Hosts that are removed during the month are charged for the full month that they are removed. Each host is treated individually. For example, when you remove a host then add a host, the bill reflects two hosts for that month.

Single-tenant instance billing
Feature	Metrics	Billing model
NFS Shared Storage at 0.25 IOPS/GB	TOTAL_STORAGE_POINT_TWO_FIVE_IOPS_GB	The peak storage allocation at the 0.25 IOPS/GB tier over the period of a month.
NFS Shared Storage at 2.0 IOPS/GB	TOTAL_STORAGE_TWO_IOPS_GB	The peak storage allocation at the 2 IOPS/GB tier over the period of a month.
NFS Shared Storage at 4.0 IOPS/GB	TOTAL_STORAGE_FOUR_IOPS_GB	The peak storage allocation at the 4 IOPS/GB tier over the period of a month.
NFS Shared Storage at 10 IOPS/GB	TOTAL_STORAGE_TEN_IOPS_GB	The peak storage allocation at the 10 IOPS/GB tier over the period of a month.

Single-tenant instance billing
Feature	Metrics	Billing model
Network edge storage		Shared edge storage with egress pricing and billed separately. Edge storage is billed monthly. Edges that are ordered after the first day of the month are prorated for the first month. Edges that are removed during the month are charged for the full month that they are removed. Each edge change is treated individually. For example, when you remove an edge then add an edge, the bill reflects both edges for that month. For an efficiency edge, one edge instance is used with 1-64 VRF (virtual edges). One efficiency edge can support up to 64 virtual data centers (VDCs) until the next edge instance is deployed. When an efficiency edge reaches attachment of zero VDCs the efficiency edge is removed.
Total egress	TOTAL_EGRESS_GB	Total outbound public networking from all virtual data centers.
Private service endpoint	MAX_PRIVATE_NETWORK_ONE_G_COST MAX_PRIVATE_NETWORK_TEN_G_COST	Monthly charge for dedicated 1 or 10 GB uplinks for the Cloud Director UI and API private network access.

Single-tenant instance billing
Feature	Metrics	Billing model
Veeam Backup service base		Monthly managed Veeam service base charge for every instance.
Veeam Backup service host		Monthly Veeam proxy storage cost for every host.
Veeam license	MAX_VEEAM_LICENSES	Veeam license charge for every VM under backup. Monthly charge for the highest number of VMs under backup at any time period in the month.
Veeam block storage	TOTAL_VEEAM_BLOCK_STORAGE_GB_HOURS	Hourly charge per GB of block storage used for all backups. Initially, all Veeam backups go to the Block Storage for Classic that is closest to their VM workloads. Backups that are a part of an inactive backup chain are immediately moved to IBM Cloud Object Storage. The restore speed for these inactive backups might be impacted. You can change how fast the inactive backup chains are moved to IBM Cloud Object Storage by opening an VCF as a Service service ticket. For more information, see Managing Veeam for VCF as a Service.
Veeam object storage	TOTAL_VEEAM_OBJECT_STORAGE_GB_HOURS	Hourly charge per GB of object storage used for all backups.

Single-tenant instance billing
Feature	Metrics	Billing model
VCDA Disaster Recovery license	DEDICATED-VCDA-DR-LICENSES	Monthly VMware Cloud Director Availability (VCDA) disaster recovery license charge for every protected VM. A VCDA VM Migration license is optionally included in your single-tenant VCF as a Service Cloud Director site order at no charge.
VCDA 20 GB private service endpoint	VCDA-PRIVATE-SERVICE-ENDPOINT	Monthly charge for every instance endpoint with a private connection.

Billing considerations for multitenant models

VCF as a Service offers two pricing plans for creating multitenant VDCs: On-demand and Reserved.

For on-demand instances, virtual data center vCPU and RAM are allocated as needed. Pricing is hourly based on the resource usage in the VDC. The amount of time that the allocation takes depends on global usage of the virtual data center vCPU and RAM. Set the maximum value limits for vCPU and RAM.

For reserved instances, the vCPU and RAM VDC reservations are pre-allocated and their availability is guaranteed. Pricing is monthly based on the allocation size of the VDC.

For both on-demand and reserved plans, you can increase and decrease the vCPU and RAM resources on a VDC later as required.

Storage quantity includes all storage that is allocated per VM, even if the VM is stopped, and all VM snapshots. Storage usage increases with new VMs and snapshots and decreases when VMs and snapshots are removed.

Storage policy availability can vary by region and deployment topology.

Multitenant instance billing
Feature	Metrics	Billing model
On-demand vCPU	TOTAL_VCPU_HOURS	The peak vCPU usage over the period of an hour. vCPU is billed hourly. vCPU quantity is for all vCPU allocated to all running VMs in the VDC at any time per hour. When a VM is stopped it is not counted in the total vCPU calculation. For example, if two running VMs both have 4 vCPU, then a third VM is started at 4 vCPU. The vCPU for the one hour period is 12 vCPU.
On-demand RAM	TOTAL_RAM_GB_HOURS	The peak memory usage over the period of an hour. \b - RAM is billed hourly. RAM quantity is for all RAM allocated to all running VMs in the VDC at any time per hour. When a VM is stopped it is not counted in the total RAM calculation. For example, if two running VMs have 16 GB RAM, then a third VM is started also at 16 GB RAM. The RAM for the one hour period is 48 GB.
Multizone On-demand vCPU		vCPU is billed hourly. vCPU quantity supports vCPU in both (2) of the data centers for all vCPU allocated to all running VMs in the VDC at any time per hour. The vCPU size is double the cost of your order to support the vCPU in both data centers. For example, if the VM has 4 vCPU and runs in a multizone region, then the total vCPU is 8. You can see the amount of vCPU that the VM uses in the primary region and IBM allocates the same amount of vCPU in the secondary region.
Multizone On-demand RAM		RAM is billed hourly. RAM quantity supports RAM in both (2) of the data centers for all RAM allocated to all running VMs in the VDC at any time per hour. The RAM size is double the cost of your order to support the RAM in both data centers. For example, if the VM has 16 GB RAM in a multizone region, then the total is 32 GB RAM. You can see the amount of RAM that the VM uses in the primary region and IBM allocates the same amount of RAM in the secondary region.

Multitenant instance billing
Feature	Metrics	Billing model
Reserved vCPU	MAX_VCPU	The peak vCPU allocation for the VDC over the period of one month. The peak vCPU metric is determined by the largest vCPU reservation value that is selected by the customer over a one month period. vCPU is billed monthly for the highest capacity reservation set for that month. VDC reservations are prorated for the first month. If the reservation is increased, then decreased again, the high water mark vCPU capacity reservation is charged for that month.
Reserved RAM	MAX_RAM_GB	The peak memory allocation for the VDC over the period of one month. The peak memory metric is determined by the largest memory reservation value that is selected by the customer over a one month period. RAM is billed monthly for the highest capacity reservation set for that month. VDC reservations are prorated for the first month. If the reservation is increased, then decreased again, the high water mark RAM capacity reservation is charged for that month.
Multizone Reserved vCPU		vCPU is billed monthly for the highest capacity reservation set for that month. The vCPU size is double the cost of your reservation to support the vCPU in both data centers. For example, if the VM has 4 vCPU reserved in a multizone region, then the total reserved vCPU is 8. You can see the amount of vCPU that the VM uses in the primary region and IBM allocates the same amount of vCPU in the secondary region.
Multizone Reserved RAM		RAM is billed monthly for the highest capacity reservation set for that month. The RAM size is double the cost of your reservation to support the RAM in both data centers. For example, if the VM has 16 GB RAM reserved in a multizone region, then the total reservation is 32 GB RAM. You can see the amount of RAM that the VM uses in the primary region and IBM allocates the same amount of RAM in the secondary region.

Multitenant instance billing
Feature	Metrics	Billing model
NFS Shared Storage at 0.25 IOPS/GB	TOTAL_STORAGE_POINT_TWO_FIVE_IOPS_GB_HOURS	The peak storage allocation at the 0.25 IOPS/GB tier over the period of an hour.
NFS Shared Storage at 2.0 IOPS/GB	TOTAL_STORAGE_TWO_IOPS_GB_HOURS	The peak storage allocation at the 2 IOPS/GB tier over the period of an hour.
NFS Shared Storage at 4.0 IOPS/GB	TOTAL_STORAGE_FOUR_IOPS_GB_HOURS	The peak storage allocation at the 4 IOPS/GB tier over the period of an hour.
NFS Shared Storage at 10 IOPS/GB	TOTAL_STORAGE_TEN_IOPS_GB_HOURS	The peak storage allocation at the 10 IOPS/GB tier over the period of an hour.
VMware vSAN storage	TOTAL_STORAGE_VSAN_GB_HOURS	The peak vSAN storage allocation over the period of an hour.
VMware standard storage		The peak standard storage allocation over the period of an hour.

Multitenant instance billing
Feature	Metrics	Billing model
Windows license	MAX_WINDOWS_LICENSES	The hourly peak number of Windows® license usage based on Windows VM vCPU size. For example, if you have two Windows VMs, one VM with 16 vCPU and one VM with 8 vCPU, the usage is 24 vCPU of Windows license usage.
Red Hat Enterprise Linux (over 127 vCPU) license	TOTAL_RHEL_LARGE_LICENSES	The hourly peak number of RHEL license usage based on VM vCPU size. One license is charged per vCPU. Large RHEL VMs are larger than 127 vCPUs.
Red Hat Enterprise Linux (9-127 vCPU) license	TOTAL_RHEL_MID_LICENSES	The hourly peak number of RHEL license usage based on VM vCPU size. One license is charged per vCPU. Mid RHEL VMs are between 9 and 127 vCPUs.
Red Hat Enterprise Linux (up to 8 vCPU) license	TOTAL_RHEL_SMALL_LICENSES	The hourly peak number of RHEL license usage based on VM vCPU size. One license is charged per vCPU. Small RHEL VMs are 8 vCPUs or smaller.

Multitenant instance billing
Feature	Metrics	Billing model
Network edge		Multiple edge storage options: shared and dedicated medium, large, or extra large Edges are billed monthly based on the type of edge used. The efficiency edge is the most cost-effective edge and the extra large edge provides the highest level of edge services and throughput over 10 Gbps. Edges are prorated for the first month. Each edge change is treated individually. For example, when you remove an edge then add an edge, the bill reflects both edges for that month. IBM recommends the Dedicated - L edge for large production workloads and high traffic. For more information, see Network edge type.
Edge gateway traffic	TOTAL_EGRESS_GB	The total outbound public traffic is charged per GB transferred over the period of one month. This value includes the outbound traffic through the VDC NSX edge to the public internet.

Multitenant instance billing
Feature	Metrics	Billing model
Veeam license	MAX_VEEAM_LICENSES	Veeam Backup service license charge for every VM under backup. Monthly charge for the highest number of VMs under backup at any time period in the month.
Veeam block storage	TOTAL_VEEAM_BLOCK_STORAGE_GB_HOURS	Hourly charge per GB of block storage used for all backups. Initially, all Veeam backups go to the Block Storage for Classic that is closest to their VM workloads. Backups that are a part of an inactive backup chain are immediately moved to IBM Cloud Object Storage. The restore speed for these inactive backups might be impacted. You can change how fast the inactive backup chains are moved to IBM Cloud Object Storage by opening an VCF as a Service service ticket. For more information, see Managing Veeam for VCF as a Service.
Veeam object storage	TOTAL_VEEAM_OBJECT_STORAGE_GB_HOURS	Hourly charge per GB of object storage used for all backups.

Multitenant instance billing
Feature	Metrics	Billing model
VCDA Disaster Recovery license	DEDICATED-VCDA-DR-LICENSES	Monthly VMware Cloud Director Availability (VCDA) disaster recovery license charge for every protected VM. A VCDA VM Migration license is optionally included in your single-tenant VCF as a Service Cloud Director site order at no charge.