What you get
A FastNode server is a full virtual machine on VMware vSphere. Your own kernel, your own init system, your own firewall rules. No container sharing a host kernel with strangers, no control panel bolted on top that you have to uninstall. You pick an image, add an SSH key, and the machine boots with you as root.
- NVMe-backed storage. Package installs, database writes and Docker layer pulls finish while you are still reading the output.
- 1 Gbps port, unmetered. No transfer allowance to watch, no overage invoice at the end of the month.
- One dedicated IPv4 per server. Set its reverse DNS yourself in the portal.
- London only. One region, one story: low round-trip times to UK users, UK ISPs and UK-hosted services.
Plans
Six sizes, one price list, no promotional first month that doubles later. Every plan includes the same network, the same images and the same portal tools. Pick by RAM first, then cores, then disk.
| Plan | vCPU | RAM | NVMe | /month | /year |
|---|---|---|---|---|---|
| Spark | 1 | 2 GB | 20 GB | $5 | $50 |
| Boost | 2 | 4 GB | 40 GB | $8 | $80 |
| Turbo | 4 | 8 GB | 80 GB | $13 | $130 |
| Nitro | 4 | 12 GB | 120 GB | $15 | $150 |
| Warp | 6 | 16 GB | 160 GB | $22 | $220 |
| Hyper | 8 | 32 GB | 200 GB | $28 | $280 |
all plans: vmware vm · 1 gbps · unmetered · 1 ipv4 · root · annual = 10 × monthly
Built for a terminal-first workflow
Most of what you do on a FastNode server happens over SSH. The portal handles the small set of jobs SSH cannot: power state, console access when networking is broken, reinstalling from a clean image, and taking a snapshot before you try something brave.
ssh [email protected]
apt update && apt -y full-upgrade
adduser deploy && usermod -aG sudo deploy
After that you're on your own box. The portal doesn't put a setup wizard in front of the shell.
The portal, briefly
Everything in the portal exists because it saves you a support ticket. The full list lives on the platform page. The ones you'll use most:
- Browser console for when you lock yourself out of SSH.
- Rescue mode to mount and repair a broken root filesystem.
- Snapshots before risky changes, and scheduled backups for everything else.
- Per-server firewall rules applied outside the guest.
- Reverse DNS editing, SSH key management, root password reset.
- CPU and bandwidth graphs so you can see what happened at 03:00.
API keys and webhooks
Create scoped API keys from your account. Each key is displayed once, at creation, and never again, so store it in your secrets manager straight away. Webhooks push account events to an endpoint you control, which means your chat channel or incident tooling can hear about changes without polling. See the API page for how keys and webhooks fit together.
Images
Seven images, kept current, nothing exotic:
- Ubuntu 24.04 LTS and Ubuntu 22.04 LTS
- Debian 12 Bookworm and Debian 13 Trixie
- AlmaLinux 9 and Rocky Linux 9
- Windows Server 2025 (talk to support about licensing before you order)
Reinstall to any of them from the portal in a couple of clicks. The disk is wiped, so snapshot first if you care about what is on it.
What people run here
The common thread is work that benefits from fast disk and short hops to UK users. Some examples, each with its own write-up:
- Self-hosted CI runners that cache dependencies on local NVMe instead of downloading them every job.
- API backends serving UK clients where every millisecond of round trip shows up in p95.
- Docker hosts running a Compose stack behind a reverse proxy.
- Databases that need real IOPS, not a burst credit that runs out.
- Staging environments that mirror production closely enough to trust.
- Self-hosted Git for teams who want their source on a box they control.
What we do not do
We are an unmanaged provider. We keep the hypervisor, network and storage healthy; you run the operating system and everything above it. We do not patch your packages, tune your database or write your Nginx config. If you want that, the guides cover the common jobs in plain steps, and the managed vs unmanaged comparison should help you decide whether you would be happier elsewhere.
We also run a single region. If your users are spread across continents you will want a multi-region provider. If they are in the UK, one well-connected London region beats five distant ones.
Start
Pick a size on the pricing page, or read about the London location first. Orders, invoices and server controls all live in the portal.