interface IEndpointInstanceProductionVariant
Language | Type name |
---|---|
.NET | Amazon.CDK.AWS.Sagemaker.Alpha.IEndpointInstanceProductionVariant |
Go | github.com/aws/aws-cdk-go/awscdksagemakeralpha/v2#IEndpointInstanceProductionVariant |
Java | software.amazon.awscdk.services.sagemaker.alpha.IEndpointInstanceProductionVariant |
Python | aws_cdk.aws_sagemaker_alpha.IEndpointInstanceProductionVariant |
TypeScript (source) | @aws-cdk/aws-sagemaker-alpha ยป IEndpointInstanceProductionVariant |
Obtainable from
Endpoint
.findInstanceProductionVariant()
Represents an instance production variant that has been associated with an endpoint.
Properties
Name | Type | Description |
---|---|---|
variant | string | The name of the production variant. |
variantName
Type:
string
The name of the production variant.
Methods
Name | Description |
---|---|
auto | Enable autoscaling for SageMaker Endpoint production variant. |
metric(namespace, metricName, props?) | Return the given named metric for Endpoint. |
metric | Metric for CPU utilization. |
metric | Metric for disk utilization. |
metric | Metric for GPU memory utilization. |
metric | Metric for GPU utilization. |
metric | Metric for the number of invocations by HTTP response code. |
metric | Metric for the number of invocations. |
metric | Metric for the number of invocations per instance. |
metric | Metric for memory utilization. |
metric | Metric for model latency. |
metric | Metric for overhead latency. |
ScaleInstanceCount(scalingProps)
autopublic autoScaleInstanceCount(scalingProps: EnableScalingProps): ScalableInstanceCount
Parameters
- scalingProps
Enable
โ EnableScalingProps.Scaling Props
Returns
Enable autoscaling for SageMaker Endpoint production variant.
metric(namespace, metricName, props?)
public metric(namespace: string, metricName: string, props?: MetricOptions): Metric
Parameters
- namespace
string
- metricName
string
- props
Metric
Options
Returns
Return the given named metric for Endpoint.
CpuUtilization(props?)
metricpublic metricCpuUtilization(props?: MetricOptions): Metric
Parameters
- props
Metric
Options
Returns
Metric for CPU utilization.
DiskUtilization(props?)
metricpublic metricDiskUtilization(props?: MetricOptions): Metric
Parameters
- props
Metric
Options
Returns
Metric for disk utilization.
GpuMemoryUtilization(props?)
metricpublic metricGpuMemoryUtilization(props?: MetricOptions): Metric
Parameters
- props
Metric
Options
Returns
Metric for GPU memory utilization.
GpuUtilization(props?)
metricpublic metricGpuUtilization(props?: MetricOptions): Metric
Parameters
- props
Metric
Options
Returns
Metric for GPU utilization.
InvocationResponseCode(responseCode, props?)
metricpublic metricInvocationResponseCode(responseCode: InvocationHttpResponseCode, props?: MetricOptions): Metric
Parameters
- responseCode
Invocation
Http Response Code - props
Metric
Options
Returns
Metric for the number of invocations by HTTP response code.
Invocations(props?)
metricpublic metricInvocations(props?: MetricOptions): Metric
Parameters
- props
Metric
Options
Returns
Metric for the number of invocations.
InvocationsPerInstance(props?)
metricpublic metricInvocationsPerInstance(props?: MetricOptions): Metric
Parameters
- props
Metric
Options
Returns
Metric for the number of invocations per instance.
MemoryUtilization(props?)
metricpublic metricMemoryUtilization(props?: MetricOptions): Metric
Parameters
- props
Metric
Options
Returns
Metric for memory utilization.
ModelLatency(props?)
metricpublic metricModelLatency(props?: MetricOptions): Metric
Parameters
- props
Metric
Options
Returns
Metric for model latency.
OverheadLatency(props?)
metricpublic metricOverheadLatency(props?: MetricOptions): Metric
Parameters
- props
Metric
Options
Returns
Metric for overhead latency.