Skip to content
SKYHIVE

VDI that users don't complain about

VDI fails in a specific way: it works in the pilot, then the login storm on the first full Monday buries it. Desktop workloads are bursty, latency-sensitive, and unforgiving, because the user is staring at every one of those milliseconds.

Honest sizing starts lower than the vendor calculators suggest: plan density conservatively, keep memory unshared, and give the storage layer the write performance a few hundred simultaneous desktops actually generate. It is cheaper to add a host later than to lose the organization's confidence in week one.

What actually determines the answer

FactorWhy it matters
Real concurrency, not seat countLicenses tell you seats; the infrastructure only feels concurrent logins. Size for the Monday-morning peak.
Desktop profile honestyA knowledge worker with two browsers and forty tabs is not a 2-vCPU task worker. Profile a real week before sizing.
Storage write latencyVDI is write-heavy and latency-sensitive. All-NVMe is the floor for a fleet that feels fast.
Failure-domain sizeWhen a host fails, everyone on it gets logged out. Smaller hosts mean smaller blast radius.

The paths, with their trade-offs

Dedicated VDI hosts

When desktop count justifies its own cluster with its own sizing rules.

Trade-off: Density per host is deliberately lower than server virtualization; resist the urge to pack it.

Configured BOM

R770 - 60 VM virtualization host

  • CPU 2× Intel Xeon 6 processor, 48 cores each
  • Memory 8× 64 GB DDR5 RDIMM (512 GB total)
  • Storage 6× 3.84 TB NVMe read-intensive SSD
  • Network Dual-port 100 GbE adapter
  • Support 4-year ProSupport Plus, next business day

Shared estate with reserved capacity

When desktop count is small and a dedicated cluster can't be justified.

Trade-off: Desktops compete with servers at the worst moments; reservations and monitoring have to be real.

Configured BOM

R770 - 30 VM virtualization host

  • CPU 2× Intel Xeon 6 processor, 32 cores each
  • Memory 8× 32 GB DDR5 RDIMM (256 GB total)
  • Storage 4× 1.92 TB NVMe read-intensive SSD
  • Network Dual-port 25 GbE adapter
  • Support 4-year ProSupport Plus, next business day

Run your own numbers first: VM sizing calculator

What deployment involves

Typical scope: host build, broker and image work with your desktop team, a measured pilot with real users, then staged rollout by department.

The pilot must include your heaviest users, not your friendliest ones.

Request a quote