You can use NFS as a secondary storage in the SDDC. When you design the physical NFS configuration, consider disk type and size, networking between the storage and the ESXi hosts, and volumes according to the data you are going to store. In this validated design, NFS stores VM templates, backups and log archives.

The management cluster uses vSAN for primary storage. When configured for a single availability zone, you use NFS for secondary storage. The shared edge and compute cluster and the compute clusters are not restricted to any particular storage technology. For compute clusters, decide on the technology for secondary storage according to the performance, capacity, and capabilities (replication, deduplication, compression, etc.) required by the workloads that are running in the clusters.

Table 1. Design Decisions on NFS Use

Decision ID

Design Decision

Design Justification

Design Implication


When using a single availability zone, provide NFS storage.

Separates primary virtual machine storage from backup data in case of primary storage failure.

vRealize Log Insight archiving requires an NFS export.

NFS storage provides the following features:

  • A datastore for backup data

  • An export for archive data

  • A datastore for templates and ISOs

An NFS capable external array is required.

You cannot configure multiple availability zones to use an NFS array in the event an availability zone fails.


When using two availability zones, store templates and ISO files on the primary vSAN datastore.

To support an availability zone failure and continue provisioning operations, the templates and ISO files must be available in the surviving availability zone.

Templates and ISO files are on the primary vSAN storage.


When using two availability zones, provide virtual machine backups in both availability zones.

To support backup and restore operations in the event of an availability zone failure, backup targets must be available in both availability zones.

The cost of the backup solution increases.