Your machine, your kernel, your problem to run. Our problem is power, cooling, uplink and the fact that out-of-band access always works.
Full access to the firmware and the metal. Install any kernel, hypervisor or weird storage stack you like — we do not audit your stack.
IPMI/KVM on every platform, reachable even when the operating system is not. That is what makes an unmanaged box survivable.
Enterprise NVMe as standard, RAID and additional disks on request. Designed for databases and caches rather than spec-sheet bravado.
| Stage | What happens |
|---|---|
| Spec | Tell us the workload. We size CPU, memory, storage and port to it — and say when a smaller box would do. |
| Quote | Hardware, bandwidth profile, term and delivery date in writing. |
| Build | Racked, cabled, firmware-checked, then handed over with credentials. |
| Handover | It is yours. We keep power, network and hardware healthy from there. |
Servers are unmanaged by default. If you want monitoring, patching or a managed layer, ask for it as an add-on rather than assume it is included.
Power, cooling, network, hardware and out-of-band access are ours. The operating system, applications, updates and backups are yours. We will fix a failed disk; we will not tune your database.
Stock platforms are usually the same working day. Custom memory, storage or ports are quoted with a delivery date rather than a vague promise.
Yes — bring your objects and we will route your prefixes to the server, or allocate from our pools with the paperwork done properly.
Usually: memory, storage, extra address space and faster ports, sometimes with a short maintenance window. If the upgrade path matters, tell us before ordering.
We monitor what we are responsible for — power, disks, temperature, link state — and contact you when a component is degrading rather than after it fails.
Ports and commits, address space, racks, or a routing question you want a second opinion on. You get an answer from an engineer, with numbers attached.
