Nutanix Snapshot Behavior with Everpure FlashArray

Nutanix Virtual Machine Storage Components

A Nutanix virtual machine consists of one or more virtual disks that store the operating system and application data. These virtual disks reside within Storage Containers that are managed by the Nutanix Distributed Storage Fabric.

When a FlashArray is connected to the Nutanix environment, the Nutanix External Storage Container represents the FlashArray inside Prism Element and Prism Central. This allows the virtual machine disks that reside on the FlashArray to be viewed and managed through the Nutanix interface.

A Nutanix virtual machine includes several disk types, each serving a different function. In addition to the primary disk for each type, Nutanix also creates a companion MetaData (MD) disk. The MD disk stores the information required for Change Block Tracking and is linked directly to the parent disk. Each disk type has its own associated MD disk.

The main Nutanix disk types are as follows:

A Nutanix virtual machine may also make use of Nutanix Volume Groups for additional storage requirements and use cases. These will be discussed later in this document.

Nutanix Controller Virtual Machine (CVM)

Nutanix node runs multiple Controller VMs (CVM), which acts as the gateway to the externally connected array. The CVM:

• Handles all I/O between Nutanix VMs and the underlying storage.

• Manages metadata, orchestration, and replication tasks across the cluster.

• Communicates with other CVMs within the cluster to maintain cluster-wide data consistency and resiliency.

• Executes API operations between the Nutanix Cluster and the Everpure FlashArray for snapshots, volume creation and other storage operations.

While every VM interacts with its local CVM for storage access, the CVM itself is a separate entity. It operates as a cluster service, abstracting and virtualizing storage resources for all workloads. The CVM requires a redundancy factor of 3 in production environments.

Managed vs. Unmanaged Snapshots

FlashArray snapshots are space efficient, thin provisioned, encrypted, deduplicated, and compressed. This allows frequent recovery points with minimal storage overhead. Their near instant creation and restoration deliver extremely low RPO and RTO without affecting production performance.

Within a Nutanix cluster connected to a FlashArray, snapshots fall into two groups: managed and unmanaged Snapshots.

Managed Snapshots (Nutanix Orchestrated FlashArray Snapshots)

Managed snapshots are initiated through Nutanix Prism Element or Prism Central. They are created on demand or through protection policies. They are also created by backup partners such as HYCU, Veeam, and others that make calls into the Nutanix APIs.

Regardless of the tool that initiates the workflow, the snapshot itself is always a FlashArray snapshot.

When Nutanix or a backup vendor initiates a snapshot through the Nutanix APIs, the following takes place.

• Nutanix calls the FlashArray snapshot APIs to create a snapshot of the VM or VM volumes and/or Nutanix volume groups.

• Nutanix records the required virtual machine and volume metadata to ensure the snapshot can be used safely for recovery and backup with CBT support.

This coordinated workflow provides the following capabilities.

• Managed snapshots appear in Prism and follow Nutanix retention and protection policies.

• They integrate fully with CBT to allow consistent incremental operations.

• They are the supported and recommended method for protecting and recovering Nutanix virtual machines. This is true whether the workflow is initiated from Prism or from a backup vendor.

In short, a managed snapshot is a FlashArray snapshot that is orchestrated by Nutanix with the required metadata tracking and lifecycle management applied by Nutanix.

Unmanaged Snapshots (Not Nutanix Aware)

Unmanaged snapshots are created directly on the FlashArray. They can be created through the Purity user interface or command line, or by features such as FlashArray Protection Groups. Nutanix does not participate in these operations.

Because Nutanix is not involved in the creation of unmanaged snapshots, the following limitations apply.

• Prism does not display, track, or manage these snapshots.

• Their presence does not create risk, but they are not suitable or supported for Nutanix virtual machine recovery.

If an administrator restores or rolls back to an unmanaged snapshot on a Nutanix VM or volume group, the following issues will occur.

• Nutanix will not have a matching metadata snapshot for that point in time, as it was not initiated by Nutanix.

• CBT will be broken, and subsequent backups will not function correctly.

• Remediation may require assistance from Nutanix Support or Pure Support, and recovery may not always be possible.

For these reasons, unmanaged snapshots should only be used for storage level workflows on the FlashArray and should not be used for Nutanix virtual machine backup or restore tasks unless managed snapshots are completely unavailable.

Future Integration Enhancements

Nutanix and Everpure are developing deeper integration with FlashArray snapshot/snapdiff APIs for a future release. This will improve coordination, visibility, and handling of array side snapshot operations. This enhancement will reduce the current limitations of unmanaged snapshots and provide a more seamless recovery and protection experience across both platforms.

Managed Snapshot Options and Consistency Types

Nutanix snapshots backed by the FlashArray are lightweight and metadata-based. They can be created manually or through scheduled protection policies in Prism Central.

Supported snapshot types include the following:

• Crash-consistent snapshots: Capture a VM's state without guest-OS coordination.

• Application-consistent snapshots: Use Nutanix Guest Tools (NGT) and Microsoft VSS (for Windows) or a custom Linux script to quiesce applications before the snapshot.

• Change Block Tracking (CBT): Tracks modified blocks between snapshots, enabling incremental backups and restores, resulting in more efficient backup jobs and data usage.

Basic Nutanix Snapshot Workflow with the FlashArray

When a Nutanix VM or vDisk resides on a Everpure FlashArray and a snapshot is taken, the following sequence results.

1. Nutanix creates a metadata snapshot in its DSF.

2. Nutanix then triggers a FlashArray snapshot of the associated volume(s) via the Pure REST API.

3. If the Nutanix snapshot is application-consistent, the FlashArray snapshot reflects that same consistency point.

4. The FlashArray snapshot serves as the physical recovery point, while Nutanix retains metadata for restoration and CBT tracking.

Mapping Nutanix Volumes to FlashArray Volumes

Everpure supports volume tagging on the FlashArray that makes it possible to identify and map vDisks and Nutanix Volume Groups between Nutanix Prism and the underlying FlashArray volumes.

Administrators can query this association with this PowerShell script. The PowerShell integration script authenticates into Nutanix and Everpure, and then it reads the Nutanix and FlashArray APIs to do the following:

• Enumerate all volumes on the FlashArray that include Nutanix tags, or optionally for a single Nutanix VM.

• Output a clear mapping table showing which FlashArray volumes belong to which Nutanix VMs and vDisks.

Nutanix Volume Groups with FlashArray

In addition to vDisks, Nutanix supports Volume Groups (VGs) that expose block volumes to VMs or external systems via iSCSI. VGs differ from vDisks because they are cluster-level objects that are not necessarily tied to a single VM. It is also important to note that a Nutanix Volume Group is not the same thing as a Everpure FlashArray Volume Groups and the two should be treated as separate entities as FlashArray Volume Groups are not supported on Nutanix today.

VGs are often used for shared-disk or database configurations across multiple VMs. Common use cases include enterprise databases such as Microsoft SQL Server, Oracle, and SAP HANA; clustered applications like Microsoft Failover Clustering and Oracle RAC.

To ensure application-consistent protection, Nutanix Consistency Groups can be used to align vDisk snapshots with Volume Group snapshots, maintaining synchronization across related data sets. This approach provides coordinated snapshot scheduling and reliable recovery points for full workload restoration.