> **Experimental**
>
> `@aws-cdk/aws-sagemaker-alpha` is an [**experimental**](/stability) (alpha) package: its API may change in any release, without a major version bump.
# IEndpointInstanceProductionVariant

Represents an instance production variant that has been associated with an endpoint.

- Stability: [Experimental](/cdk/api/stability.html)
- Source: [`lib/endpoint.ts`](https://github.com/aws/aws-cdk/blob/v2.275.0-alpha.0/packages/@aws-cdk/aws-sagemaker-alpha/lib/endpoint.ts#L55)

## Properties

- `variantName`: string — The name of the production variant.

## Methods

- `autoScaleInstanceCount(scalingProps: EnableScalingProps): ScalableInstanceCount` — Enable autoscaling for SageMaker Endpoint production variant.

  **Parameters**

  - `scalingProps` — EnableScalingProps.

- `metric(namespace: string, metricName: string, props?: MetricOptions): Metric` — Return the given named metric for Endpoint.

- `metricCpuUtilization(props?: MetricOptions): Metric` — Metric for CPU utilization.

- `metricDiskUtilization(props?: MetricOptions): Metric` — Metric for disk utilization.

- `metricGpuMemoryUtilization(props?: MetricOptions): Metric` — Metric for GPU memory utilization.

- `metricGpuUtilization(props?: MetricOptions): Metric` — Metric for GPU utilization.

- `metricInvocationResponseCode(responseCode: InvocationHttpResponseCode, props?: MetricOptions): Metric` — Metric for the number of invocations by HTTP response code.

- `metricInvocations(props?: MetricOptions): Metric` — Metric for the number of invocations.

- `metricInvocationsPerInstance(props?: MetricOptions): Metric` — Metric for the number of invocations per instance.

- `metricMemoryUtilization(props?: MetricOptions): Metric` — Metric for memory utilization.

- `metricModelLatency(props?: MetricOptions): Metric` — Metric for model latency.

- `metricOverheadLatency(props?: MetricOptions): Metric` — Metric for overhead latency.