Autoscaling

class verta.endpoint.autoscaling.Autoscaling(min_replicas=None, max_replicas=None, min_scale=None, max_scale=None)

Represents autoscaling configuration for Endpoint, to be passed to Endpoint.update().

The JSON equivalent for this is:

{
    "autoscaling": {
        "quantities": {"min_replicas": 2, "max_replicas": 7, "min_scale": 0.2, "max_scale": 1.3},
        "metrics": []
    }
}
Parameters:
  • min_replicas (int) – Minimum number of replicas to scale down to.

  • max_replicas (int) – Maximum number of replicas to scale up to.

  • min_scale (float in (0, 1)) – Minimum growth factor for scaling.

  • max_scale (float > 1) – Maximum growth factor for scaling.

Examples

from verta.endpoint.autoscaling import Autoscaling
autoscaling = Autoscaling(min_replicas=1, max_replicas=10, min_scale=0.5, max_scale=2.0)
add_metric(metric)

Adds a metric.

Parameters:

metric (metrics) – Metric to add.