Virtualization
Virtualization that survives losing a host
VMware vSphere and Proxmox/KVM clusters, shared storage, HA, migration between platforms and tested recovery — not just the install, but the daily operation.
The problem
A cluster works right up until you lose the first host
Standing virtualization up is easy; what happens afterwards is not. Licences expire, storage turns out to be the single point of failure, nobody tests the backup by restoring it, host versions drift apart, and live migration fails exactly when it is needed.
- Shared storage is a single point of failure
- Backups exist; a restore has never been tested
- Host versions and configuration have drifted
- Nobody has counted capacity — N+1 exists only on paper
What we deliver
Virtualization that survives losing a host
Cluster design
Host count, N+1, network segmentation and storage topology.
Storage and its resilience
ZFS/RAID levels, NFS/iSCSI, snapshots and replication.
HA and live migration
The lost-host scenario tested, not assumed.
Backup and recovery
Policy, retention and a restore rehearsal on a schedule.
Migration
Physical to virtual, or VMware to Proxmox, in stages.
Multi-site management
Several clusters in several locations — one view, one procedure, one backup policy.
Documentation and handover
Runbooks, diagrams and training for your own team.
Architecture
Architecture
- 01Physical
- Redundant power
- Cooling
- OOB
- 02NetworkLeaf-spine
- EVPN-VXLAN
- LACP
- MTU 9000
- 03ComputeVirtualization cluster
- KVM / Proxmox
- VMware
- HA
- 04Storage
- Ceph
- ZFS
- NVMe
- 05RecoveryTested restore
- Snapshots
- Off-site
- DR runbook
Operator consoles
Operator consoles
These exact systems run on the group's own infrastructure — the screenshots are processed before publication.
The virtualization cluster
A multi-node Proxmox cluster: KVM virtual machines and LXC containers, NFS/HA storage, a separate backup NAS and migrated ESXi hosts. This website runs on it too — in a container, on the same infrastructure we sell.
8 clusters from one interface
Proxmox Datacenter Manager brings independent Proxmox VE clusters into one place. In our own infrastructure it manages 8 of them, spread across separate physical locations — 3 of those outside Georgia. The frame also shows the scale: 211 nodes online, 1,480 virtual machines and 127 containers — one view, one procedure.
vSphere — the cluster, its hosts and the workloads on it
A vCenter with several ESXi hosts and the services placed on them: web, billing, a domain controller, backup agents. This is the environment where VMware and Proxmox coexist — part of it still on vSphere, part already migrated. One team operates both, under one backup policy.
Storage — ZFS pools and replication
ZFS storage on a RAIDZ2 topology — 12 disks in one vdev, roughly 33 TiB usable, ECC memory with most of it serving as ZFS cache. And the part that matters: disks with errors, zero; scrub and scan, clean. This is the layer underneath the virtualization cluster and the backups.
Backup and recovery
Centralised backup for virtual and physical systems: protection status per machine, the storage usage trend, and a separate list of hosts that stopped reporting. That last list is what matters — a backup nobody watches is not a backup.
Capabilities
Capabilities
Every item is marked: verified production experience, or engineering capability.
VMware vSphere / ESXi
ProvenvCenter, a multi-host cluster, shared storage and workload placement — operated daily inside the group's own infrastructure.
- vSphere
- ESXi
- vCenter
Proxmox VE and KVM/LXC
ProvenA multi-node cluster, NFS/HA storage, containers and virtual machines on one platform.
- Proxmox VE
- KVM
- LXC
Multi-cluster management (Proxmox Datacenter Manager)
ProvenIndependent Proxmox VE clusters through one interface: node state, virtual machines and containers, CPU/RAM/storage and backup servers. In our own infrastructure it manages 8 clusters across separate physical locations — 3 of them outside Georgia.
- Datacenter Manager
- multi-cluster
- PBS
VMware to Proxmox migration
ProvenA staged move where both platforms coexist until the last workload lands — the licence expiry does not get to set the schedule.
- ESXi import
- qm importovf
- rsync
ZFS storage and replication
ProvenRAIDZ pools, snapshot schedules and disk health monitoring.
- TrueNAS
- ZFS
- NFS
Backup and tested restore
ProvenCentralised backup with agents, a retention policy and a restore rehearsal — without the rehearsal it is not a backup.
- Cyber Backup
- ZFS snapshots
HA and automatic restart
CapabilityWorkloads brought back on another node when a host is lost.
- HA cluster
- Fencing
Provisioning automation
CapabilityTemplates, cloud-init and the API — a hand-built VM should be the exception.
- Terraform
- Ansible
- cloud-init
Process
Process
- 01
Audit
Existing hosts, storage, licences and workloads.
- 02
Design
Target cluster, N+1 and storage topology.
- 03
Build
Nodes, network, storage and backup — documented.
- 04
Migrate
In stages, with a rollback plan and agreed windows.
- 05
Validate
Losing a host and restoring from backup are actually performed, not assumed.
- 06
Operate
Monitoring, capacity and the patch cycle.
Technology stack
Technology stack
- Hypervisor
- VMware vSphereESXiProxmox VEKVMLXC
- Management
- Proxmox Datacenter ManagervCenterProxmox Backup Server
- Storage
- TrueNASZFSNFSiSCSIRAID 10 SSD
- Backup
- Cyber BackupZFS snapshotsReplication
- Automation
- AnsibleTerraformcloud-init
- Monitoring
- ZabbixPrometheusGrafanaLibreNMS
Engagement model
Engagement model
Project
A one-off scope: audit, migration or implementation with a fixed outcome.
Retainer
Monthly engineering hours — specialist access on demand.
Co-managed
NetWizard and your in-house team together, with split responsibility.
Fully managed
We own the agreed operational scope end to end.
Use cases
Use cases
The licence is running out
The VMware renewal got more expensive — part of the estate moves to Proxmox, part stays, and one team operates both.
One host, no spare
Everything runs on one server. We build a cluster with shared storage, where losing a host is an incident rather than a disaster.
A backup nobody has restored
We set the policy, rehearse the restore into a scratch environment and record the result in writing.
FAQ
FAQ
VMware or Proxmox?
It depends on the existing workloads, the licence budget and your team's experience. We operate both in production and owe nothing to either vendor — the recommendation comes after the audit, not before it.
Does migration mean downtime?
Most workloads move inside an agreed window; critical services move by replication and a short cutover. Every step carries a rollback plan.
Do you really test restores?
Yes — the restore runs into a separate database or a scratch environment and the result is recorded. Until that is done, we treat the backup status as unknown.