vSAN Ready Nodes or Hyper-V and a SAN for a host refresh?

November 16, 2026 5 answers
Asked by 2k~ticketspindle_481

We're planning a refresh for a five-node VxRail cluster running about 55 VMs, mostly Windows. The current cluster has 53.5 TiB of storage.

A vendor proposed four vSAN Ready Nodes with 256 GB RAM and 256 GB NVMe for memory tiering per host, or five Hyper-V hosts with 512 GB RAM each and a traditional SAN. Is that memory difference realistic? What should we compare for licensing, resilience, and room to grow?

Answers

8j!foldpocket_264

That is a big sizing gap, and I wouldn't accept hypervisor overhead as the whole explanation without seeing the calculations. Hyper-V has dynamic memory too. Ask the vendor to show the assumed VM memory demand, host failure allowance, maintenance headroom, growth, and storage requirements for both designs.

Start with actual memory usage and peaks, not just the number of VMs or their configured maximums. Then check whether the remaining hosts can carry the workload after a failure, including whatever growth and maintenance scenarios you need to cover.

NVMe memory tiering can help, but it isn't interchangeable with RAM. Its value depends on the workload and how much memory is actively used. Ask for workload-specific testing or sizing assumptions before relying on it to close that much of a gap.

2w~bitcradle_948:

I'd also ask whether the four-node design still meets its performance target during a host failure, not just whether the VMs can start.

Reply to this answer
_z52!parcelnotch.3

For the license comparison, get written quotes for the whole stack rather than assuming the Windows-heavy workload makes either option cheaper. Include Windows Server rights, VMware subscriptions, management tools, support, and the physical core counts in each design.

Windows Server Datacenter licensing may already cover Hyper-V on properly licensed hosts, while System Center Virtual Machine Manager licensing depends on what you've purchased. Some Core Infrastructure Suite licensing includes System Center. Check the exact agreement and current terms with your reseller, including how VM mobility affects licensing.

Reply to this answer
r.284!nodefunnel

Separate the hypervisor decision from the storage decision. Hyper-V with a SAN is one option, but Hyper-V with Storage Spaces Direct is another. On the VMware side, vSAN Ready Nodes keep storage distributed across the hosts rather than in a separate array.

Compare usable capacity after the resilience settings, storage performance during failures, network requirements, and what happens when you need to expand. With a SAN, check the array, controllers, connectivity, and support costs as well as the hosts. Also look at where that 53.5 TiB is going. A few large VMs can change the sizing conversation quite a bit.

Reply to this answer
_k81+willowratchet.6

What are the actual uptime and recovery requirements? That should drive how much redundancy you buy. A hyperconverged cluster might be a good fit, but don't assume you need one just because the current system uses it.

Local storage may be worth considering if the workload and recovery targets allow it. Compare the operational tradeoffs, including how you would move VMs for maintenance and recover them after a host or disk failure.

Reply to this answer
_b64+hingequartz.2

I'd get at least one quote outside those two designs. Nutanix is another hyperconverged option, and Proxmox is worth evaluating if its management and support model fit your team. There are also Hyper-V-based options beyond a traditional SAN.

Make vendors show migration effort, ongoing support, and renewal costs alongside the initial price. If the existing hardware still has useful support life, check whether extending or reusing it makes sense too. No point refreshing more than necessary.

Reply to this answer

Add your answer

Answers are reviewed before publication.

Related Discussions