However, one specific capability, the Number of Failures to Tolerate (FTT), has a significant effect on storage capacity and therefore cost, so it is addressed here. For more information on assigning storage polices in vSAN refer to: http://www.vmware.com/content/dam/digitalmarketing/vmware/en/pdf/vcat/architecting-vmware-virtual-san-62-for-vmware-vcloud-air-network.pdf

The Number of Failures to Tolerate capability addresses the key customer and design requirement of availability. With FTT, availability is provided by maintaining replica copies of data, to mitigate the risk of a host failure resulting in lost connectivity to data or potential data loss. The FTT policy works in conjunction with VMware vSphere High Availability to maintain availability and provide consistent and near continuous uptime to workloads.

In its most basic form, the FTT policy provides mirrored (RAID1) availability. For instance, a FTT=1 supports n+1 availability by providing a second copy of data on a separate host in the cluster. However, as you would expect, the resulting impact on capacity is that it is doubled. For example, if you deploy a virtual machine with a 100 GB disk, and copy 20 GB of data to the VM, without any further configuration of space efficiency technologies, the following impact on sizing would result:

However, as mentioned, this is FTT in its most basic form. If you employ space efficiency technologies, such as erasure coding and deduplication and compressions in an all-flash configuration, erasure coding will yield at least a 33 percent space saving, and deduplication and compression, which is more difficult to calculate because it is dependent on workload type, could yield another 30-40 percent saving. the use of these space efficiency technologies in an all-flash configuration is one of the reasons why the all-flash value proposition is so compelling. Also, consider that all VMDK objects are by default sparse (thin), with no performance benefit of pre-allocating space.

vSAN license metering for VMware Cloud Providers is calculated based on consumed capacity of the vSAN datastores, irrespective of all other factors, such as the number of hosts in the cluster, number of VMs, or type of data stored. The metered value on which the cost is calculated is based on consumed datastore capacity, which is derived by subtracting freeCapacityB from totalCapacityB, leaving us with Used Capacity. As a result, the FTT policy not only has a significant impact on capacity sizing, but also has an impact on OpEx-based VMware Cloud Provider Program licensing costs.

Does this mean that VMware Cloud Providers pay for replica copies of data provided for redundancy? Yes, although it is not quite as straightforward as that from a cost comparison perspective with other solutions, because the following storage capacity efficiency technologies have a big impact on the Used Capacity being reported:

Therefore, while the Number of Failures to Tolerate policy must be considered when designing and preparing a business case and total cost of ownership analysis for vSAN, a VMware Cloud Provider will not be doubling or tripling the storage costs for replica data when these space efficiency technologies are taken into account. 

For more information on storage policies, including the number of failures to tolerate, erasure coding and deduplication and compression, please refer to: http://www.vmware.com/content/dam/digitalmarketing/vmware/en/pdf/vcat/architecting-vmware-virtual-san-62-for-vmware-vcloud-air-network.pdf