In the server room of a hundred-person furniture manufacturer in Kayseri stand four boxes aged four to nine years. One holds the Logo Tiger SQL database, another is the terminal server where dealers key in orders and home workers log on, a third keeps production drawings. The fourth is an old desktop nobody dares switch off, because the USB licence dongle for the CAM software that writes CNC programs is plugged into it. None of them uses even half its capacity, yet each is a link that can snap on its own. A burnt-out power supply in the terminal server stops dealer orders; a dead disk in the file server stalls the production plan. Virtualisation turns these servers into virtual machines that can move between two hosts. If one host fails, the machines start on the other, and a snapshot taken before patching makes rollback a matter of minutes. Planning, build and migration are handled remotely. You purchase hardware from your own supplier, who, or a local firm, takes care of racking and cabling.
The costliest virtualisation mistakes are made before the purchase order, not during installation. Most of the scope is therefore about answering questions.
A USB licence dongle, a weighbridge or gauge on a serial port, legacy software tied to one particular network card. Before anything is ordered, we confirm with your software vendor whether such devices can be passed through to a virtual machine.
A week of measuring processor, memory, disk and network load, month-end peaks included. Add growth headroom and spare capacity for the day one host has to run everything by itself.
Hyper-V where the estate is all Windows, Proxmox VE with ZFS and Ceph where cost and flexibility come first, VMware where a software vendor requires it. You receive a written comparison of strengths and weaknesses for your situation.
Windows Server Standard grants two virtual Windows instances once every physical core on the host is licensed, whereas Datacenter grants unlimited instances. We calculate which is cheaper for your planned VM count and produce a three-year total including platform and backup software.
Replicated local SSDs, shared storage, or a Ceph cluster from three hosts upwards. Management, replication and application traffic each travel on their own network.
Every server moves in an agreed slot and is checked by the staff who rely on its application daily. For the first few days a path back to the old hardware remains open.
At the end of the project we deliberately shut down a host and time how quickly, and in which order, the machines come up on the other one. The results are written into the recovery plan.
How long it takes mainly reflects the number of servers and hardware lead times. None of the work requires us on site.
Inventory, measurements, dependencies, and a discussion with management about how long each system can afford to be down.
Platform choice, a hardware list you can pass to your supplier, network plan, backup and recovery design, plus the three-year total cost.
After delivery and racking by your local firm, remote installation, network and storage configuration, and performance testing.
Servers move one by one, the failure drill is run, and the recovery plan is handed to you.
A single-host set-up concentrates all your risk in one chassis. Collapsing four old servers into one powerful machine is tempting and the invoice stays small. When that lone host dies, however, all four workloads stop at once. If the budget cannot cover two hosts, we discuss it openly and at minimum put in place fast-restoring backups, a pre-quoted replacement hardware plan and a scenario for briefly running critical machines in the cloud. Whatever risk you choose to carry is recorded in writing.
Usually a few hours per server, scheduled outside business hours and often over a weekend. Database servers such as Logo are kept well clear of month-end closing. Which server goes offline, when and for how long is announced in writing beforehand.
Yes, provided someone who knows it is in charge. Proxmox VE is open source, and its developer sells paid subscriptions with a stable update repository and vendor support. We recommend that subscription for production use.
While your agreement runs, there is no hurry. In light of VMware's recent licensing changes, we price a renewal against Hyper-V or Proxmox and check whether your backup tool and business applications support the alternative. Moving can happen gradually.
Possibly. Virtual machines can equally run at European providers such as Hetzner, OVHcloud or IONOS, or in EU regions of Azure and AWS. Whether that makes sense hinges on data volumes, your software's licence terms, special hardware like USB dongles and your internet connection. We cost both options side by side.
The manufacturer under warranty, or your supplier. Replication keeps the virtual machines running on the surviving host meanwhile. We manage the warranty process and return the repaired host to service.
Which servers do you run today, what sits on them, and is any special equipment attached? We will reply with our first questions and an outline design.
We have your enquiry
A reply will reach you by the next working day at the latest. If your message says work has come to a halt, it goes to the top of the pile.
No such city in our list. Try another spelling, or just pick the closest big city: we work entirely over remote connections, so nothing about the service changes from one province to the next.
The only cookies here are the essential ones: they keep the site running and remember the city you picked. Nothing is used for advertising or tracking. See our privacy notice for more.
