Your own compute. Managed around the work.
Stack Servers give your business a private pool of CPU, memory and storage — managed by StackHost.
Use the capacity for one large server or several smaller workloads. We shape the environment around what your applications need today and expand it when the work changes.
Live regions Dallas · DenverCustom pools quoted on request
Capacity you reserve, machines we shape around it.
The pool is the product.
Traditional server plans make every workload fit a predefined machine size. Stack Servers work the other way round.
Reserve the CPU, memory and storage your environment needs, then divide that capacity into the machines and services the workload actually requires.
One pool can contain one machine or several.
Example workloads inside one pool
A staging environment can sit beside production in the same pool. So can the queue that feeds it. Neither has to wait for a new server order.
Stop paying for the wrong-shaped server.
Workloads are not the same shape. A fixed machine size asks every one of them to pretend otherwise.
A private compute pool lets the environment be shaped around those differences.
One workload needs more memory
It runs comfortably while its CPU allocation sits mostly idle.
Another needs more storage
Uploads, logs and backup retention grow faster than traffic does.
A third barely uses CPU
It runs quietly most of the day and should not pay for cores it never touches.
Then the balance changes
The pool is re-shaped around the new workload mix instead of the whole environment being lifted and moved.
Private infrastructure without running it alone.
Reserved capacity is the starting point. What makes it a managed lane is everything StackHost operates around it.
One question usually decides it.
Both lanes can carry serious production work. The difference is who owns the operating layer, and how many machines one order can contain.
“I want to manage the server.”
You hold root and run the operating system yourself.
- Operating system
- Packages
- Applications
- Patching
- Security inside the VM
“I want the infrastructure, managed.”
StackHost manages the agreed environment while you focus on the application and the business requirements around it.
- Private reserved pool
- One or several machines
- Environment operated for you
- Monitoring and backups
- Capacity you can grow into
Starting shapes, not hard limits.
S, M and L are published so the lane is easy to scan. A pool is then shaped around the workloads it runs.
| Starting pool | CPU | Memory | Storage | From | Usually shaped into |
|---|---|---|---|---|---|
| Stack Server S | 2 vCPU | 4 GB | 80 GB | $47.45 / mo | One application that has outgrown a shared plan. |
| Stack Server M | 4 vCPU | 8 GB | 160 GB | $88.47 / mo | An app plus its database, or production beside staging. |
| Stack Server L | 8 vCPU | 16 GB | 320 GB | $170.51 / mo | Several services a part of the business depends on at once. |
| Custom pool | Your number | Your number | Your number | Quoted | Whatever balance the workload actually asks for. |
These are useful starting points. A Stack Server can be configured differently when the workload needs another balance of CPU, memory and storage — tell us what runs and we will say whether a listed size is the cheaper answer.
A backup gives you a copy. Recovery gives you a way back.
They are sold separately because they answer different problems, and a heavy workload may need one, both, or neither.
A copy you can restore from
For the ordinary operational mistakes: an accidental deletion, a bad update, a broken deployment, damaged data.
Scheduled per the agreed retention, and tested when you ask us to.
A separate path back
For the larger failure a backup alone will not cover — an environment lost, not a file lost.
Designed and priced around the workload. Failover is planned with you. Not automatic.
Managed compute should remove work and show where your responsibility starts.
Unlike a Cloud VPS, a Stack Server holds capacity for you — and you do not have to turn the environment around it into a second job. Routine pool operations, monitoring and standard maintenance are included; hands-on troubleshooting, recovery and custom work draw on your intervention allowance or support balance.
Physical placement, the virtualisation layer, host-side patching, the network path, and the monitoring that tells us something is wrong before your customers do.
The hosting environment itself: how the machines inside the pool are shaped, web server and runtime configuration, certificate renewals, database provisioning, queue upkeep and the agreed backup schedule.
The applications, their data, how they behave under load, what you deploy to them, and the growth decisions that eventually make the current pool too small.
When something needs attention, the ring it sits in decides who moves first — which is why incidents on this lane tend to be short conversations rather than long ones.
Your workload does not have to come in S, M or L.
The starting sizes exist because hosting should be easy to scan. The stronger part of the offer is a private pool shaped around the workload — CPU, memory and storage set per machine.