Skip to content

Autoscaling

There is nothing to configure. Ownkube sizes each web app and worker automatically, within your plan’s size per service: 1 vCPU and 1 GB on Pay as you go, 8 vCPU and 8 GB on Personal, 8 vCPU and 16 GB on Team. You pay per minute for what the app uses, with no billed minimum.

Compute doesn’t add or remove copies for you, so horizontal autoscaling isn’t offered there. To run more than one copy, set the replica count on the deployment’s Settings tab, up to your plan’s replicas per service: 1 on Pay as you go, 6 on Personal, 20 on Team. See how the limits count.

Autoscaling adds replicas to a web app or worker under load and removes them when load drops. Once it’s on, Ownkube owns the replica count. Turn it on from the deployment’s Settings tab on your own cloud.

Ownkube watches the average CPU of a deployment’s containers against the target you set, as a percentage of the CPU request. When usage crosses the target Ownkube adds replicas up to your maximum; when usage falls it removes them down to your minimum. This is the horizontal autoscaling model: more copies of the same container, not a bigger container.

CPU target

Scale out when average CPU exceeds the target, for example 70% of the CPU request.

Memory target

Scale out when average memory exceeds the target, for example 80% of the memory request.

Set one target, both, or neither. With both, whichever is crossed first wins.

Field Default Range
Min replicas 1 1 to 100
Max replicas 3 1 to 100
Target CPU utilization 80% 1 to 100
Target memory utilization 80% 1 to 100

For a web service on your own cloud with traffic that varies through the day:

Setting Value
Min replicas 2
Max replicas 10
Target CPU utilization 70%
CPU request 1000m
Memory request 1Gi

Two replicas stay warm, up to ten run under load (scaling out once they average 700 millicores), and the count settles back down when traffic drops.

  • Once autoscaling is on, the fixed Replicas field is ignored. Turn autoscaling off to set a manual count.
  • Autoscaling and sleep are exclusive. Sleep needs a single replica; autoscaling needs at least one always on.
  • Jobs run to completion and do not scale. Databases scale by changing their box size, see Postgres.

On either kind of cluster, the Overview tab shows running and desired replicas and the CPU and memory charts the scaler reads. Scale events appear on the Activity tab. A CPU or memory alert at a lower threshold than the scaling target warns you before the maximum is reached.