Skip to content

PredictiveScalingMetricSpecification

Source code in docs_src/capo_auto_scaling/types/predictive_scaling_metric_specification.py
class PredictiveScalingMetricSpecification(TypedDict, closed=True):
    target_value: NotRequired["capo_auto_scaling.types.metric_scale.MetricScale"]
    """<p>Specifies the target utilization.</p> <note> <p>Some metrics are based on a count instead of a percentage, such as the request count for an Application Load Balancer or the number of messages in an SQS queue. If the scaling policy specifies one of these metrics, specify the target utilization as the optimal average request or message count per instance during any one-minute interval. </p> </note>"""
    predefined_metric_pair_specification: NotRequired[
        "capo_auto_scaling.types.predictive_scaling_predefined_metric_pair.PredictiveScalingPredefinedMetricPair"
    ]
    """<p>The predefined metric pair specification from which Amazon EC2 Auto Scaling determines the appropriate scaling metric and load metric to use.</p>"""
    predefined_scaling_metric_specification: NotRequired[
        "capo_auto_scaling.types.predictive_scaling_predefined_scaling_metric.PredictiveScalingPredefinedScalingMetric"
    ]
    """<p>The predefined scaling metric specification.</p>"""
    predefined_load_metric_specification: NotRequired[
        "capo_auto_scaling.types.predictive_scaling_predefined_load_metric.PredictiveScalingPredefinedLoadMetric"
    ]
    """<p>The predefined load metric specification.</p>"""
    customized_scaling_metric_specification: NotRequired[
        "capo_auto_scaling.types.predictive_scaling_customized_scaling_metric.PredictiveScalingCustomizedScalingMetric"
    ]
    """<p>The customized scaling metric specification.</p>"""
    customized_load_metric_specification: NotRequired[
        "capo_auto_scaling.types.predictive_scaling_customized_load_metric.PredictiveScalingCustomizedLoadMetric"
    ]
    """<p>The customized load metric specification.</p>"""
    customized_capacity_metric_specification: NotRequired[
        "capo_auto_scaling.types.predictive_scaling_customized_capacity_metric.PredictiveScalingCustomizedCapacityMetric"
    ]
    """<p>The customized capacity metric specification.</p>"""

customized_capacity_metric_specification instance-attribute

customized_capacity_metric_specification: NotRequired[
    PredictiveScalingCustomizedCapacityMetric
]

The customized capacity metric specification.

customized_load_metric_specification instance-attribute

customized_load_metric_specification: NotRequired[PredictiveScalingCustomizedLoadMetric]

The customized load metric specification.

customized_scaling_metric_specification instance-attribute

customized_scaling_metric_specification: NotRequired[
    PredictiveScalingCustomizedScalingMetric
]

The customized scaling metric specification.

predefined_load_metric_specification instance-attribute

predefined_load_metric_specification: NotRequired[PredictiveScalingPredefinedLoadMetric]

The predefined load metric specification.

predefined_metric_pair_specification instance-attribute

predefined_metric_pair_specification: NotRequired[PredictiveScalingPredefinedMetricPair]

The predefined metric pair specification from which Amazon EC2 Auto Scaling determines the appropriate scaling metric and load metric to use.

predefined_scaling_metric_specification instance-attribute

predefined_scaling_metric_specification: NotRequired[
    PredictiveScalingPredefinedScalingMetric
]

The predefined scaling metric specification.

target_value instance-attribute

target_value: NotRequired[MetricScale]

Specifies the target utilization.

Some metrics are based on a count instead of a percentage, such as the request count for an Application Load Balancer or the number of messages in an SQS queue. If the scaling policy specifies one of these metrics, specify the target utilization as the optimal average request or message count per instance during any one-minute interval.