In storage administration, "flexibility" is often the hardest requirement to satisfy. You want high availability, but you also need architectural freedom. You want disaster recovery, but you need it to fit within specific budget and hardware constraints. You want massive scale, but you want to start small and grow organically.
Nutanix Unified Storage (NUS) 5.3 is defined by its ability to deliver on these needs without the traditional tradeoffs. It decouples the capabilities you rely on from the infrastructure constraints that usually govern them. Whether it’s streamlining Kubernetes® provisioning, enhancing metadata scalability, or bringing analytics to air-gapped environments, NUS 5.3 is about giving you the freedom to build the architecture that best fits your business.
Untethered Disaster Recovery and Availability
One of the most significant shifts in this release is how we handle replication topologies. Historically, Smart DR—our native file replication engine—required the source and target clusters to be symmetrical. While robust, this often meant that a large production cluster required an equally large disaster recovery site, regardless of the actual performance needs at the secondary location.
With NUS 5.3, we introduce Asymmetric Smart DR. By updating the underlying replication engine to a file-centric model, we have effectively decoupled the source topology from the target. You can now replicate a large, performance-heavy production cluster to a smaller, capacity-dense target cluster. This allows for a much more cost-effective DR strategy, allowing you to only provision the compute and storage resources you actually need at your recovery site.
Something to note: As this leverages an updated replication engine, this capability applies to new Smart DR policies created after upgrading.
We have also expanded flexibility for mission-critical Metro Availability (RPO 0) deployments on AHV. Previously, managing a Metro setup required a single Prism Central (PC) instance to oversee both sites. Now, NUS 5.3 supports Dual-PC Files Metro, allowing you to maintain independent Prism Central instances at both the active and standby sites, coordinated by a standalone Witness. This enhances failure domain isolation and aligns better with geographically distributed operations.
Accelerating Modern Workloads
Kubernetes environments are dynamic by nature, often creating and destroying persistent volumes at a pace that traditional storage struggles to match. NUS 5.3 introduces Containerized Mode for Nutanix Files Storage, designed specifically to bridge this gap.
In this mode, the file server "pre-warms" the backing infrastructure required for new shares. Instead of spinning up resources on demand—which can introduce latency during high-churn CI/CD cycles—the system maintains a ready-to-go pool of resources. This optimization significantly reduces the time it takes to provision new shares, ensuring that storage is ready effectively the moment your Kubernetes clusters ask for it.
We see similar efficiency gains in High Availability (HA). For File Servers running on AHV, we have refined the detection logic for failover events. Rather than relying solely on standard network timeouts to confirm a node failure, the system now references the Acropolis Power State directly. If the hypervisor reports that a VM has powered off, Nutanix Files can bypass the traditional "wait and verify" checks and initiate recovery immediately, leading to a much faster restoration of service.
Granular Control and Security
As file environments grow, so does the need for precision. Administrators have long sought better ways to manage capacity within complex directory structures. In the past, limiting the size of a specific project folder often required workarounds like creating connected shares just to apply a quota.
NUS 5.3 introduces true Directory Level Quotas, allowing you to apply hard or soft limits directly to specific paths within a share. This simplifies capacity management and helps prevent "share sprawl" by removing the need to create new shares solely for administrative control.
Security also gets an upgrade with native support for NFSv4 ACLs. While standard mode bits provide a reliable foundation, modern enterprises often require the fine-grained inheritance and specific allow/deny rules found in complex permission environments. With NFSv4 ACL support, NUS allows for a unified security posture that honors granular permission structures regardless of the protocol being used.
Nutanix Objects Storage: Built to Scale
Nutanix Objects Storage continues to mature as a platform for massive, data-heavy workloads. A key enhancement in 5.3 is the re-architecture of the metadata layer to run on Blockstore, the same high-performance engine powering our core block storage.
This shift unlocks enhanced Metadata Scale-Out and Scale-Up. You can now add Metadata Servers to an existing object store as a seamless Day-2 operation, allowing the system's performance to grow organically alongside your data. Furthermore, the efficiency of this new architecture allows each metadata node to handle significantly more objects than before, keeping your infrastructure efficient even as you scale into billions of objects.
For our analytics-driven customers, Nutanix Objects Storage is now fully validated with Dremio. This allows you to run high-performance SQL queries directly against data sitting in Nutanix Objects (S3). When paired with the Nutanix Kubernetes Platform (NKP) to host the Dremio engine (also recently validated), you have a powerful, on-premise Data Lakehouse solution.
Streamlining Operations and Upgrades
Operational simplicity remains a core tenet of NUS. For customers operating in dark sites or air-gapped environments, 5.3 introduces Direct Upload for Objects upgrades. This removes the need for external web servers, allowing admins to upload the LCM bundle—containing all necessary container images—directly to Prism Central via the browser.
This capability is powered by image pre-fetching, which also significantly accelerates upgrades for standard, internet-connected clusters. Previously, the system fetched upgrade images only after the object store went offline. Now, images are staged locally before the maintenance window even opens. Once the upgrade begins, the system operates more efficiently by upgrading almost all microservices simultaneously rather than serially. This combination of pre-fetching and parallelization provides for maintenance windows that are faster and less disruptive than ever, allowing S3 clients to tolerate the brief outage with appropriate timeout and retry settings.
Nutanix Data Lens Comes Home
Finally, for organizations with strict data sovereignty requirements that preclude the use of SaaS platforms, we are introducing Data Lens On-Premises. This new deployment option brings the core value of our analytics service—including ransomware visibility and audit trails—directly to your Prism Central cluster.
While the SaaS version continues to offer the broadest feature set—at least for the time being—the on-premises version ensures that federal, defense, and highly regulated customers can now benefit from deep data visibility while keeping all metadata strictly behind their firewall.
Other Notable Enhancements
- Secure Authentication: LDAPS (LDAP over SSL) support for communication with Active Directory ensures compliance with modern security standards.
- Smart DR Config Sync: Share settings and quotas now automatically replicate to the DR site, so your recovery environment is always in sync with production.
- High Performance Deployment: Deploying the "High Performance" Files profile is now fully integrated into the UI and API, simplifying setup for demanding workloads.
- Cloud Tiering Expansion: Official support for Google Cloud Storage and OVHcloud Object Storage as S3 Smart Tiering targets.
- Advanced Networking: Flow Advanced VLANs are now supported for both client and storage networks on PE-deployed File Servers.
- Federation Control: Custom Bucket Placement allows specific control over which Objects cluster within an Objects global namespace owns specific data.
- Event Notifications: Kafka integration has been improved with support for mTLS and multiple endpoints per topic.
- Accessibility: Significant updates to A11y standards across the UI for a more inclusive experience.
©2026 Nutanix, Inc. All rights reserved. Nutanix, the Nutanix logo and all Nutanix product and service names mentioned are registered trademarks or trademarks of Nutanix, Inc. in the United States and other countries. Kubernetes is a registered trademark of The Linux Foundation in the United States and other countries. All other brand names mentioned are for identification purposes only and may be the trademarks of their respective holder(s).