Share via


IOnlineDeploymentProperties Interface

Definition

[System.ComponentModel.TypeConverter(typeof(Microsoft.Azure.PowerShell.Cmdlets.MachineLearningServices.Models.Api20240401.OnlineDeploymentPropertiesTypeConverter))]
public interface IOnlineDeploymentProperties : Microsoft.Azure.PowerShell.Cmdlets.MachineLearningServices.Models.Api20240401.IEndpointDeploymentPropertiesBase
[<System.ComponentModel.TypeConverter(typeof(Microsoft.Azure.PowerShell.Cmdlets.MachineLearningServices.Models.Api20240401.OnlineDeploymentPropertiesTypeConverter))>]
type IOnlineDeploymentProperties = interface
    interface IJsonSerializable
    interface IEndpointDeploymentPropertiesBase
Public Interface IOnlineDeploymentProperties
Implements IEndpointDeploymentPropertiesBase
Derived
Attributes
Implements

Properties

AppInsightsEnabled

If true, enables Application Insights logging.

CodeConfigurationCodeId

ARM resource ID of the code asset.

(Inherited from IEndpointDeploymentPropertiesBase)
CodeConfigurationScoringScript

[Required] The script to execute on startup. eg. "score.py"

(Inherited from IEndpointDeploymentPropertiesBase)
DataCollectorCollection

[Required] The collection configuration. Each collection has it own configuration to collect model data and the name of collection can be arbitrary string. Model data collector can be used for either payload logging or custom logging or both of them. Collection request and response are reserved for payload logging, others are for custom logging.

DataCollectorRollingRate

When model data is collected to blob storage, we need to roll the data to different path to avoid logging all of them in a single blob file. If the rolling rate is hour, all data will be collected in the blob path /yyyy/MM/dd/HH/. If it's day, all data will be collected in blob path /yyyy/MM/dd/. The other benefit of rolling path is that model monitoring ui is able to select a time range of data very quickly.

Description

Description of the endpoint deployment.

(Inherited from IEndpointDeploymentPropertiesBase)
EgressPublicNetworkAccess

If Enabled, allow egress public network access. If Disabled, this will create secure egress. Default: Enabled.

EndpointComputeType

[Required] The compute type of the endpoint.

EnvironmentId

ARM resource ID or AssetId of the environment specification for the endpoint deployment.

(Inherited from IEndpointDeploymentPropertiesBase)
EnvironmentVariable

Environment variables configuration for the deployment.

(Inherited from IEndpointDeploymentPropertiesBase)
InstanceType

Compute instance type.

LivenessProbeFailureThreshold

The number of failures to allow before returning an unhealthy status.

LivenessProbeInitialDelay

The delay before the first probe in ISO 8601 format.

LivenessProbePeriod

The length of time between probes in ISO 8601 format.

LivenessProbeSuccessThreshold

The number of successful probes before returning a healthy status.

LivenessProbeTimeout

The probe timeout in ISO 8601 format.

Model

The URI path to the model.

ModelMountPath

The path to mount the model in custom container.

Property

Property dictionary. Properties can be added, but not removed or altered.

(Inherited from IEndpointDeploymentPropertiesBase)
ProvisioningState

Provisioning state for the endpoint deployment.

ReadinessProbeFailureThreshold

The number of failures to allow before returning an unhealthy status.

ReadinessProbeInitialDelay

The delay before the first probe in ISO 8601 format.

ReadinessProbePeriod

The length of time between probes in ISO 8601 format.

ReadinessProbeSuccessThreshold

The number of successful probes before returning a healthy status.

ReadinessProbeTimeout

The probe timeout in ISO 8601 format.

RequestLoggingCaptureHeader

For payload logging, we only collect payload by default. If customers also want to collect the specified headers, they can set them in captureHeaders so that backend will collect those headers along with payload.

RequestSettingMaxConcurrentRequestsPerInstance

The number of maximum concurrent requests per node allowed per deployment. Defaults to 1.

RequestSettingMaxQueueWait

(Deprecated for Managed Online Endpoints) The maximum amount of time a request will stay in the queue in ISO 8601 format. Defaults to 500ms. (Now increase request_timeout_ms to account for any networking/queue delays)

RequestSettingRequestTimeout

The scoring timeout in ISO 8601 format. Defaults to 5000ms.

ScaleSettingScaleType

[Required] Type of deployment scaling algorithm

Methods

ToJson(JsonObject, SerializationMode) (Inherited from IJsonSerializable)

Applies to