Autoscaling

Autoscaling your Deployment to adapt to changes in Deployment metrics.

Configure autoscaling to handle scale up and down events for your Gradient deployment.

How autoscaling works

Gradient autoscaling uses the kubernetes horizontal pod autoscaler. Some defaults have been chosen to make it easier to quickly scale up and down the deployment.

Autoscaling will scale up and down the deployment based on a chosen metric , summary function and specified value. The number of current replicas for each deployment will never scale below replicas or above maxReplicas.

Scaledown is calculated on a 5 minute period. This means that if your application is underutilized for 5 minutes, it will scale down to the number of replicas required to handle the current load.

To change the autoscaling configuration, update the spec through the Paperspace console or CLI.

Autoscaling configuration

enabled (default: true): Turn autoscaling on or off.

maxReplicas : The upper bound on the number of replicas that can be run by the deployment. The deployment’s active replicas will always fall in the range between the value of replicas and maxReplicas.

metric - Sets the metric used to scale up or down.

summary - Sets the function used to calculate scale events.

value - The summary number that will cause the deployment to scale.

Autoscaling criteria

Multiple metrics can be used in the spec to determine when to scale. If you provide multiple metric blocks, the deployment will calculate a proposed replica counts for each metric, and then scale the instances to the value of the highest replica count.

The following metrics can be used:

metric	summary	Description	Type
cpu	average	Average cpu utilization across each replica (% of 100)	Integer
memory	average	Average memory utilization across each replica (% of 100)	Integer
requestDuration	average	Average request duration over a 5 minute period across all IPs behind the proxy (seconds). Minimum of 10 milliseconds	Float

Autoscaling example

A spec that configures all metrics available for autoscaling. See example scenario below:

resources:
  replicas: 1
  ...
  autoscaling:
    enabled: true # toggle for enabling/disabling autoscaling
    maxReplicas: 3 # max replicas for autoscaling
    metrics:
      - metric: cpu
        summary: average
        value: 50 # 50% cpu utilization across all replicas
      - metric: memory
        summary: average
        value: 22 # 22% memory utilization across all replicas
      - metric: requestDuration
        summary: average
        value: 0.15 # 150 millisecond request duration for the endpoint

Example Scenario: As requests begin to come through, the request duration over a 5 minute period is greater than 150 ms. As a result, the deployment scales up from 1 to 2 replicas. Over the next 5 minute interval the request duration is still longer than 150ms and the deployment scales to 3 replicas. After the stabilization period of 5 minutes, the deployment will begin to scale down as the request times have fallen below 150 ms.