Compute
Virtual machines on demand. Provision the capacity a workload needs, resize it when the workload changes, and release it when it is no longer required.
Information
Compute is the service that turns UNIQCloud capacity into virtual machines your teams control. You choose a size, an operating system image, and a network, and the platform delivers a running server in minutes. UNICC operates the underlying hypervisors, the physical hosts, and the three Availability Zones the workloads are spread across; you keep full administrative control of what runs inside the virtual machine.
Because capacity is billed as it is consumed, Compute changes the economics of running infrastructure: there is no procurement cycle in front of a new project, and no stranded hardware behind a decommissioned one.
Resize as demand changes
A machine can be moved to a larger size as a service grows. Capacity follows the workload instead of being committed up front for its whole lifetime.
Spread across three zones
Compute capacity is available in all three Availability Zones of the Valencia data centre, so an application can be designed to survive the loss of a zone.
Your machine, your rules
UNICC operates the platform; you retain root access, patching policy, and software choice inside the instance. Access is controlled per project and per user.
Virtual machine sizes
Compute is offered as a catalogue of fixed sizes, called flavors. Each flavor is a set combination of vCPU, memory, and root disk; choosing one is how you declare the capacity — and therefore the cost — of a machine. Sizes can be changed later by resizing the instance.
| Specification | cpu1-ram2 disk16 | cpu2-ram4 disk32 | cpu2-ram8 disk32 | cpu4-ram8 disk32 | cpu4-ram16 disk32 | cpu8-ram16 disk32 | cpu16-ram32 disk32 |
|---|---|---|---|---|---|---|---|
| vCPU | 1 | 2 | 2 | 4 | 4 | 8 | 16 |
| Memory | 2 GB | 4 GB | 8 GB | 8 GB | 16 GB | 16 GB | 32 GB |
| Root disk (ephemeral) | 16 GB | 32 GB | 32 GB | 32 GB | 32 GB | 32 GB | 32 GB |
| Typically used for | Small services, jump hosts, agents | Light web and application servers | Memory-sensitive small services | Standard production application servers | Application servers and small databases | CPU-bound processing, cluster nodes | Large databases, batch and analytical jobs |
All root and ephemeral disks in these flavors are ephemeral, meaning data is lost when the VM is terminated. For production workloads, we strongly recommend using persistent volumes with Cinder to ensure data durability and reliability.
For more details on setting up persistent storage, see our guide: Block Storage Volumes.
Project quotas
Every project is created with a set of default resource quotas. They exist to guarantee fair usage across tenants, keep the platform stable, and make capacity planning possible — and, because Managed IaaS is billed on consumption, they are also the main control available for keeping a project's spend predictable.
Compute
| Resource Type | Quota |
|---|---|
| Instances | 20 |
| VCPUs | 30 |
| RAM | 50 GB |
Storage
| Resource Type | Quota |
|---|---|
| Volumes | 30 |
| Volume Snapshots | 20 |
| Volume Storage | 1.5 TB |
Networking
| Resource Type | Quota |
|---|---|
| Floating IPs | 50 |
| Security Groups | 10 |
| Security Rules | 100 |
| Networks | 10 |
| Ports | 50 |
| Routers | 10 |
If your workload requires more resources, quotas can be adjusted on request. For quota increases, please contact support with:
- Project name
- Required resources
- Expected duration and use case
Usage
Day-to-day operation of virtual machines — creating, resizing and deleting instances, reaching a console, and managing the SSH keys used to log in.