Coming from VMware?
Your Windows Server and ERP virtual machines run on Cloud Hypervisor, under one licence for the whole installation, priced per core. Most of what your team knows already has a name here.
| In VMware | In Velrix |
|---|---|
| vCenter | Manager, the operators’ window |
| ESXi host | Host: each server runs the host module |
| Virtual machine | Machine, on Cloud Hypervisor |
| VM template | Blueprint, in the operator’s catalogue |
| Port group | Network, per project |
| Roles and permissions | Roles per organisation |
| Cluster | Host pool, with its own failover and maintenance rules |
| vSphere HA | Availability policy per app and per host pool |
| Maintenance mode | Drain a host |
| vMotion | Live migration |
| Moving your VMs | Import from vCenter, in waves with rollback |
| Veeam backups | Backups of the platform, volumes and VM disks |
When a host goes silent, stateless containers start on another host after about 90 seconds, keeping their addresses. A VM whose disk is on that host waits for it and operators are told; Velrix never deletes that disk on its own.
What moving over looks like
Import from vCenter
A planning report maps each vCenter VM to a host pool, size and network, with its monthly cost. VMs move in waves that roll back to their untouched source.
Availability policies
Each app sets its restarts, its health checks, and what happens when its host is lost or cut off, within the limits its host pool and the installation allow.
Drain for maintenance
Cordon and drain a host from the Manager. The drain honours each app’s disruption budget, moves VMs live, and shows its progress per app as it goes.
Live migration
A running VM moves to another host in its pool, disk included, with a pause of about 100 ms in our tests. A guest that will not settle is slowed until it does (auto-converge).
Backups
Scheduled, encrypted backups of platform state, volumes and VM disks, frozen through the guest agent for a consistent copy. Restore in place or as a copy.
VMM updates
When a new VM runtime ships, each existing VM shows it and takes it on its next restart, in a maintenance window you choose. Its disk moves onto a volume once, checksum-verified.
Also on the platform: logs with live tail per deployment, OpenTofu and Ansible runs with approvals, and updates in your own maintenance windows.
See Velrixon your own terms
We walk through the platform and the architecture, and how Velrix would run in your organisation.