ComputeOptimizer / Client / put_recommendation_preferences

put_recommendation_preferences#

ComputeOptimizer.Client.put_recommendation_preferences(**kwargs)#

Creates a new recommendation preference or updates an existing recommendation preference, such as enhanced infrastructure metrics.

For more information, see Activating enhanced infrastructure metrics in the Compute Optimizer User Guide.

See also: AWS API Documentation

Request Syntax

response = client.put_recommendation_preferences(
    resourceType='Ec2Instance'|'AutoScalingGroup'|'EbsVolume'|'LambdaFunction'|'NotApplicable'|'EcsService'|'License'|'RdsDBInstance'|'Idle',
    scope={
        'name': 'Organization'|'AccountId'|'ResourceArn',
        'value': 'string'
    },
    enhancedInfrastructureMetrics='Active'|'Inactive',
    inferredWorkloadTypes='Active'|'Inactive',
    externalMetricsPreference={
        'source': 'Datadog'|'Dynatrace'|'NewRelic'|'Instana'
    },
    lookBackPeriod='DAYS_14'|'DAYS_32'|'DAYS_93',
    utilizationPreferences=[
        {
            'metricName': 'CpuUtilization'|'MemoryUtilization',
            'metricParameters': {
                'threshold': 'P90'|'P95'|'P99_5',
                'headroom': 'PERCENT_30'|'PERCENT_20'|'PERCENT_10'|'PERCENT_0'
            }
        },
    ],
    preferredResources=[
        {
            'name': 'Ec2InstanceTypes',
            'includeList': [
                'string',
            ],
            'excludeList': [
                'string',
            ]
        },
    ],
    savingsEstimationMode='AfterDiscounts'|'BeforeDiscounts'
)
Parameters:

  • resourceType (string) –

    [REQUIRED]

    The target resource type of the recommendation preference to create.

    The Ec2Instance option encompasses standalone instances and instances that are part of Auto Scaling groups. The AutoScalingGroup option encompasses only instances that are part of an Auto Scaling group.

  • scope (dict) –

    An object that describes the scope of the recommendation preference to create.

    You can create recommendation preferences at the organization level (for management accounts of an organization only), account level, and resource level. For more information, see Activating enhanced infrastructure metrics in the Compute Optimizer User Guide.

    Note

    You cannot create recommendation preferences for Auto Scaling groups at the organization and account levels. You can create recommendation preferences for Auto Scaling groups only at the resource level by specifying a scope name of ResourceArn and a scope value of the Auto Scaling group Amazon Resource Name (ARN). This will configure the preference for all instances that are part of the specified Auto Scaling group. You also cannot create recommendation preferences at the resource level for instances that are part of an Auto Scaling group. You can create recommendation preferences at the resource level only for standalone instances.

    • name (string) –

      The name of the scope.

      The following scopes are possible:

      • Organization - Specifies that the recommendation preference applies at the organization level, for all member accounts of an organization.

      • AccountId - Specifies that the recommendation preference applies at the account level, for all resources of a given resource type in an account.

      • ResourceArn - Specifies that the recommendation preference applies at the individual resource level.

    • value (string) –

      The value of the scope.

      If you specified the name of the scope as:

      • Organization - The value must be ALL_ACCOUNTS.

      • AccountId - The value must be a 12-digit Amazon Web Services account ID.

      • ResourceArn - The value must be the Amazon Resource Name (ARN) of an EC2 instance or an Auto Scaling group.

      Only EC2 instance and Auto Scaling group ARNs are currently supported.

  • enhancedInfrastructureMetrics (string) –

    The status of the enhanced infrastructure metrics recommendation preference to create or update.

    Specify the Active status to activate the preference, or specify Inactive to deactivate the preference.

    For more information, see Enhanced infrastructure metrics in the Compute Optimizer User Guide.

  • inferredWorkloadTypes (string) –

    The status of the inferred workload types recommendation preference to create or update.

    Note

    The inferred workload type feature is active by default. To deactivate it, create a recommendation preference.

    Specify the Inactive status to deactivate the feature, or specify Active to activate it.

    For more information, see Inferred workload types in the Compute Optimizer User Guide.

  • externalMetricsPreference (dict) –

    The provider of the external metrics recommendation preference to create or update.

    Specify a valid provider in the source field to activate the preference. To delete this preference, see the DeleteRecommendationPreferences action.

    This preference can only be set for the Ec2Instance resource type.

    For more information, see External metrics ingestion in the Compute Optimizer User Guide.

    • source (string) –

      Contains the source options for external metrics preferences.

  • lookBackPeriod (string) –

    The preference to control the number of days the utilization metrics of the Amazon Web Services resource are analyzed. When this preference isn’t specified, we use the default value DAYS_14.

    You can only set this preference for the Amazon EC2 instance and Auto Scaling group resource types.

    Note

    • Amazon EC2 instance lookback preferences can be set at the organization, account, and resource levels.

    • Auto Scaling group lookback preferences can only be set at the resource level.

  • utilizationPreferences (list) –

    The preference to control the resource’s CPU utilization threshold, CPU utilization headroom, and memory utilization headroom. When this preference isn’t specified, we use the following default values.

    CPU utilization:

    • P99_5 for threshold

    • PERCENT_20 for headroom

    Memory utilization:

    • PERCENT_20 for headroom

    Note

    • You can only set CPU and memory utilization preferences for the Amazon EC2 instance resource type.

    • The threshold setting isn’t available for memory utilization.

    • (dict) –

      The preference to control the resource’s CPU utilization threshold, CPU utilization headroom, and memory utilization headroom.

      Note

      This preference is only available for the Amazon EC2 instance resource type.

      • metricName (string) –

        The name of the resource utilization metric name to customize.

      • metricParameters (dict) –

        The parameters to set when customizing the resource utilization thresholds.

        • threshold (string) –

          The threshold value used for the specified metric parameter.

          Note

          You can only specify the threshold value for CPU utilization.

        • headroom (string) –

          The headroom value in percentage used for the specified metric parameter.

          The following lists the valid values for CPU and memory utilization.

          • CPU utilization: PERCENT_30 | PERCENT_20 | PERCENT_0

          • Memory utilization: PERCENT_30 | PERCENT_20 | PERCENT_10

  • preferredResources (list) –

    The preference to control which resource type values are considered when generating rightsizing recommendations. You can specify this preference as a combination of include and exclude lists. You must specify either an includeList or excludeList. If the preference is an empty set of resource type values, an error occurs.

    Note

    You can only set this preference for the Amazon EC2 instance and Auto Scaling group resource types.

    • (dict) –

      The preference to control which resource type values are considered when generating rightsizing recommendations. You can specify this preference as a combination of include and exclude lists. You must specify either an includeList or excludeList. If the preference is an empty set of resource type values, an error occurs. For more information, see Rightsizing recommendation preferences in the Compute Optimizer User Guide.

      Note

      • This preference is only available for the Amazon EC2 instance and Auto Scaling group resource types.

      • Compute Optimizer only supports the customization of Ec2InstanceTypes.

      • name (string) –

        The type of preferred resource to customize.

        Note

        Compute Optimizer only supports the customization of Ec2InstanceTypes.

      • includeList (list) –

        The preferred resource type values to include in the recommendation candidates. You can specify the exact resource type value, such as m5.large, or use wild card expressions, such as m5. If this isn’t specified, all supported resources are included by default. You can specify up to 1000 values in this list.

        • (string) –

      • excludeList (list) –

        The preferred resource type values to exclude from the recommendation candidates. If this isn’t specified, all supported resources are included by default. You can specify up to 1000 values in this list.

        • (string) –

  • savingsEstimationMode (string) –

    The status of the savings estimation mode preference to create or update.

    Specify the AfterDiscounts status to activate the preference, or specify BeforeDiscounts to deactivate the preference.

    Only the account manager or delegated administrator of your organization can activate this preference.

    For more information, see Savings estimation mode in the Compute Optimizer User Guide.

Return type:

dict

Returns:

Response Syntax

{}

Response Structure

  • (dict) –

Exceptions