ray.serve.config.AutoscalingConfig#

class ray.serve.config.AutoscalingConfig(*, min_replicas: NonNegativeInt = 1, initial_replicas: NonNegativeInt | None = None, max_replicas: PositiveInt = 1, target_num_ongoing_requests_per_replica: PositiveFloat = 1.0, target_ongoing_requests: PositiveFloat | None = None, metrics_interval_s: PositiveFloat = 10.0, look_back_period_s: PositiveFloat = 30.0, smoothing_factor: PositiveFloat = 1.0, upscale_smoothing_factor: PositiveFloat | None = None, downscale_smoothing_factor: PositiveFloat | None = None, upscaling_factor: PositiveFloat | None = None, downscaling_factor: PositiveFloat | None = None, downscale_delay_s: NonNegativeFloat = 600.0, upscale_delay_s: NonNegativeFloat = 30.0)[source]#

Bases: BaseModel

Config for the Serve Autoscaler.

Methods

`construct`	Creates a new model setting __dict__ and __fields_set__ from trusted or pre-validated data.
`copy`	Duplicate a model, optionally choose which fields to include, exclude and change.
`dict`	Generate a dictionary representation of the model, optionally specifying which fields to include or exclude.
`get_policy`	Deserialize policy from cloudpickled bytes.
`json`	Generate a JSON representation of the model, `include` and `exclude` arguments as per `dict()`.
`serialize_policy`	Serialize policy with cloudpickle.
`update_forward_refs`	Try to update ForwardRefs on fields based on this Model, globalns and localns.

Attributes

`min_replicas`
`initial_replicas`
`max_replicas`
`target_num_ongoing_requests_per_replica`
`target_ongoing_requests`
`metrics_interval_s`
`look_back_period_s`
`smoothing_factor`
`upscale_smoothing_factor`
`downscale_smoothing_factor`
`upscaling_factor`
`downscaling_factor`
`downscale_delay_s`
`upscale_delay_s`