Customers running Dell VxRail infrastructure now have an official path to upgrade their existing systems to VMware Cloud Foundation (VCF) 9.1 without wiping and reinstalling the hosts or necessarily replacing the hardware. Dell Technologies and VMware jointly developed the procedure, which also introduces an important twist for future expansion: certain environments will be able to combine VxRail clusters with Dell vSAN Ready Nodes within the same VCF instance.
VxRail and VMware Cloud Foundation 9.1 in 30 seconds
- Dell and VMware have built an in-place upgrade path from VxRail to VCF 9.1 for compatible configurations.
- There’s no need to wipe, reinstall, or re-image VxRail hosts during the process.
- VMware 5.2.2 or later and VxRail 8.0.361 at minimum are required before starting the upgrade.
- VCF 9.1 allows VxRail and Dell vSAN Ready Nodes to be combined under specific conditions.
- Stretch Clusters, Dynamic Nodes, two-node configurations, and Satellite Nodes will have to wait for a later release.
The announcement matters because it changes the options available to companies that made significant investments in VxRail hyperconverged infrastructure and are now working out how to fit those systems into the new generation of VMware Cloud Foundation.
The alternative no longer necessarily involves refreshing hardware to move to VCF 9.1. Systems that meet the stated requirements can follow the standard VxRail upgrade procedure and keep their existing servers.
An upgrade for the VxRail you already have installed
The most significant feature of the announcement is probably summed up in a single word VMware keeps using: in-place.
Dell and VMware jointly developed a path that avoids a repave — that is, wiping and recreating the hosts on VxRail appliances.
There are prerequisites.
Before starting the migration, the environment’s VMware software must be on version 5.2.2 or later, while VxRail must already be updated to at least version 8.0.361. From there, administrators can use the established VxRail upgrade process to move to VCF 9.1.
This could be especially important for deployments running on servers that still have useful life left.
Refreshing a hyperconverged platform doesn’t just affect server pricing. It can mean capacity planning, migrations, maintenance windows, networking and storage changes, testing, and engineering hours.
A supported upgrade on existing hardware makes it possible to separate two decisions that don’t always have to happen at the same time: upgrading the software platform and refreshing the physical infrastructure.
That doesn’t mean any VxRail can be upgraded. Configurations still have to stay within the hardware and support boundaries validated by Dell.
There are also architectures that remain out of scope for now.
Customers running Stretch Clusters, Dynamic Nodes, two-node configurations, or Satellite Nodes will have to wait for a release after VCF 9.1 to get a supported upgrade path. VMware doesn’t give a date for that support in the announcement.
VCF 9.1 starts to break hardware homogeneity
The second piece of news could turn out to be even more interesting in the medium term.
VMware Cloud Foundation deployments have traditionally relied on a high degree of infrastructure homogeneity. A VxRail-based deployment kept growing on VxRail, while a vSAN Ready Nodes-based one followed the other model.
VCF 9.1 partially loosens that separation.
A VCF instance running on VxRail can bring in new workload domains built on Dell vSAN Ready Nodes. Organizations can keep their existing VxRail clusters and turn to another compatible Dell architecture whenever they need to add capacity.
That doesn’t mean VMware allows freely mixing any certified server.
The restrictions are fairly specific.
Within each Workload Domain, the hardware still has to be homogeneous: the domain must be made up entirely of VxRail or entirely of Dell vSAN Ready Nodes.
In addition, whenever VxRail systems exist within the VCF instance, the Management Domain must remain exclusively on VxRail.
Third-party vSAN Ready Nodes aren’t supported within a VCF instance running on VxRail, either. The ability to mix hardware is limited to the model validated and supported by Dell.
So it would be inaccurate to describe VCF 9.1 as a platform that’s fully hardware-independent.
There’s more freedom than before, but within clearly defined boundaries.
Two lifecycle models under one VCF
Managing different hardware within the same platform also raises an operational question: how to update each cluster.
VCF 9.1 keeps different mechanisms depending on the underlying infrastructure.
VxRail clusters continue to use their built-in Lifecycle Management (LCM), while clusters built on Dell vSAN Ready Nodes use VCF’s standard lifecycle process.
Above both, VCF Operations and SDDC Manager provide the management plane in charge of coordinating the different compute, storage, and networking components.
VCF continues to rely on technologies like vSphere, vSAN, and NSX to do so.
The architecture makes it possible to keep VxRail’s operational specifics without forcing every future expansion to use exactly the same type of node.
For organizations with a large installed base, this could become a more gradual way to evolve their infrastructure.
A data center could keep its VxRail clusters for as long as they have enough capacity and support, upgrade the platform to VCF 9.1, and later add workload domains on Dell vSAN Ready Nodes as new needs come up.
VxRail isn’t going away, but its role is changing
The collaboration also partly clears up questions some customers may have had about VxRail’s future within VMware Cloud Foundation.
VMware explicitly states that VCF on VxRail remains a Dell-exclusive solution, designed and supported by the manufacturer.
At the same time, VCF 9.1 is trying to offer platform capabilities more consistently across the different supported hardware architectures.
The result is a model somewhat different from the one that originally characterized hyperconvergence.
VxRail keeps its specific integration, but no longer forces every new workload domain in that instance to necessarily use VxRail.
For infrastructure teams, this widens their planning options.
A company can decide to stick with VxRail, gradually bring in Ready Nodes, or look at entirely different alternatives once it’s time to refresh its platform. Having an upgrade path to VCF 9.1 means that decision no longer has to be made purely because the software needs a version bump.
It also buys time to weigh another unavoidable factor: the total cost of staying on VMware Cloud Foundation.
Being able to keep the hardware doesn’t automatically mean the upgrade is the best economic option for every customer. Licensing, support, capacity needs, remaining equipment lifespan, and virtualization strategy will keep shaping each case.
One more option for those reconsidering VMware
The announcement also lands at a time when many organizations are reviewing their virtualization strategy and evaluating platforms such as Proxmox VE, Nutanix AHV, Hyper-V, or OpenStack for certain scenarios.
For a VxRail customer, that evaluation has a particular wrinkle: there’s already an investment in infrastructure designed around VMware.
A supported path to VCF 9.1 lets that investment keep paying off for longer and eases the pressure to migrate right away.
But it doesn’t remove the reasons that might lead a company to look at other platforms, either.
The decision will depend on licensing costs, required features, technology lock-in, automation, in-house skills, and the expected timeline for refreshing servers.
What the Dell and VMware announcement does change is the number of options.
Owners of compatible VxRail systems no longer necessarily have to frame the situation as an immediate choice between replacing infrastructure and abandoning the platform.
There’s now a third way: upgrade the existing VxRail to VCF 9.1, keep it running for as long as it remains fit for purpose, and decide later how to evolve new workload domains.
That can give infrastructure leaders with hundreds of thousands of euros still invested in perfectly functional VxRail systems considerably more breathing room.
Frequently asked questions
Can an existing VxRail be upgraded to VMware Cloud Foundation 9.1?
Yes, for compatible configurations. Dell and VMware have built an official in-place upgrade path that requires VMware 5.2.2 or later and VxRail 8.0.361 at minimum beforehand.
Do VxRail servers need to be wiped or reinstalled?
Not on the announced path. VMware states the procedure doesn’t require a repave — that is, wiping and re-imaging the VxRail hosts.
Does VCF 9.1 let you mix VxRail with other servers?
It allows combining VxRail clusters and Dell vSAN Ready Nodes within the same instance under the conditions set by Dell. Third-party vSAN Ready Nodes aren’t supported in a VCF instance running on VxRail, and each Workload Domain must use homogeneous hardware.
Can every VxRail environment be upgraded already?
No. Stretch Clusters, Dynamic Nodes, two-node configurations, and Satellite Nodes will have to wait for a release after VCF 9.1 for a supported path.
