One vCPU, one thread. Yours.
Each vCPU is pinned to a physical thread that carries nothing else. Not oversubscribed, not time-sliced, not shared at 4 a.m. either.
Your vCPU are pinned to your machine and no other guest is ever scheduled on them. Same building, same jurisdiction, none of the variance.
Size your server
Core 8
3 tiers
| Plan | vCPU | RAM | NVMe | Transfer | Port | Ready in | Price | Order |
|---|---|---|---|---|---|---|---|---|
| Core 4 Databases, indexers, anything latency-sensitive | 4 dedicated | 16 GB | 1 TB | Unlimited | 10 Gbps | ~60 s | $39$31/mo | Deploy Core 4 |
| Core 8 Busy production stacks, CI runners | 8 dedicated | 32 GB | 2 TB | Unlimited | 10 Gbps | ~60 s | $75$60/mo | Deploy Core 8 |
| Core 16 Analytics, large caches, media pipelines | 16 dedicated | 64 GB | 4 TB | Unlimited | 10 Gbps | ~60 s | $139$111/mo | Deploy Core 16 |
Every tier: pinned vCPU · 1 IPv4 · routed IPv6 /64 · always-on DDoS filtering · 10 Gbps port · full API. Need more cores than the top row? Ask — we build them.
What pinned means
Shared cores are not slow — they are unpredictable. The number you measure on a Sunday is not the number you get on a Tuesday, and no amount of tuning fixes a neighbour compiling something.
One vCPU, one thread. Yours.
Each vCPU is pinned to a physical thread that carries nothing else. Not oversubscribed, not time-sliced, not shared at 4 a.m. either.
No steal time. Ever.
Run vmstat and the steal column stays at zero. It is the one number that tells you whether a host is telling the truth.
Ten-gigabit port from the floor.
The entry tier already gets the 10 Gbps port that the shared line only reaches three rows up. Reserved cores are wasted behind a throttled link.
Twice the disk of the shared line.
One terabyte where the equivalent shared tier gives 480 GB. Workloads that need predictable CPU usually need somewhere to put the data.
No upsells
No protected-IP tier, no per-snapshot charge, no support plan to buy.
Full root, KVM
Real hardware virtualisation. Your own kernel, Docker, nested VMs, sysctl. Not a container wearing a server costume.
IPv4 + a routed /64
One IPv4 on every plan, more on request, and a whole /64 of IPv6 so you never subnet a /128.
DDoS filtering
Layer 3/4 scrubbing upstream of the port, on by default, at no charge and with nothing to enable.
Two snapshot slots
Take one before you break something. Restore in under a minute, as often as you like.
Custom ISO and console
Mount any image, including your own hardened build, and watch it boot over the serial console.
API and SSH keys
Provision, resize, snapshot and destroy from your own code. Keys installed at order time, password login off.
Before you order
Anything else, support answers in minutes — and says no plainly when the answer is no.
Open a ticketThe hardware is still shared — other guests run on the same machine — but never on your cores or your memory. A dedicated server gives you the whole chassis, IPMI and a longer provisioning time. This resizes in a minute; that does not.
Ask support for a two-hour trial. We hand you a live instance, you run whatever you like on it, and you keep the numbers. No card, nothing to cancel.
Yes, and support does the migration: rsync while the old server keeps running, a short cutover, then the old disk is wiped. No charge, and no downtime worth the name.
Yes. Memory is allocated, not ballooned — the host never reclaims pages from you to give them to somebody else, which is the other half of why shared instances vary.
The rest of the range
Same jurisdiction, same network, same refusal to ask who you are. They differ on the one thing named in each title.
Ready when you are
No account needed to see every price, no identity needed to order. $39 a month gets you a server nobody can talk us into switching off.