Built for the edge and the mid-market Broadcom walked away from.
The same firmware runs a two-node shop and a five-hundred-node datacenter. Here's where teams put it to work.
Edge & distributed sites
Two- and three-node HA clusters in stores, branches, and remote sites, each streaming back to a central datacenter. Survive a site outage with central failover.
Industrial & on-prem
Air-gapped clusters on the factory floor with no phone-home and no kill-switch. Deterministic, sovereign infrastructure that runs whether or not the internet does.
Public sector & sovereign
Runs with no internet connection at all: local sign-in, an on-site control plane, and signed updates you import yourself. Nothing leaves the perimeter.
AI / GPU workloads
GPU passthrough with cluster-aware placement and no per-GPU software tax. Fill dense accelerator boxes without watching the bill climb.
VMware migration
Connect to an ESXi host, inventory the VMs, and migrate them in with automated disk conversion and minimal downtime, with a credit covering your existing contract.
Service providers
Monthly per-node billing with no minimum term and built-in multitenancy. Usage reported automatically, so you bill your tenants, not your cores.
300 stores, one image, zero truck-rolls.
A retailer runs a two-node Fabric cluster in every store, each streaming sales and inventory data back to the regional datacenter. When a store's hardware fails, its workloads restart centrally from the last signed restore point while a replacement node ships.
- A replacement node is discovered, placed and joined automatically — no on-site configuration.
- Signed OTA updates roll the fleet without sending a technician on-site.
- Small store boxes sit in the Edge class at $500/node/yr, whatever you fill them with.
Your workload not listed?
If it runs as a VM or a container, it runs on Fabric. Start free on a single node and see.