Organiza tus páginas con colecciones
Guarda y categoriza el contenido según tus preferencias.
En esta página, se describe cómo ver y cambiar la configuración de tus recomendaciones de tipos de máquinas. Puedes modificar los tipos de máquinas que se recomiendan para que las recomendaciones generadas solo incluyan las series de máquinas que prefieras.
También puedes cambiar las métricas que se usan para generar las recomendaciones de memoria y, así, mejorar su precisión. Los cambios en la configuración se aplican a nivel del proyecto.
Si aún no lo hiciste, configura la autenticación.
La autenticación verifica tu identidad para acceder a los servicios y las APIs de Google Cloud . Para ejecutar código o muestras desde un entorno de desarrollo local, puedes autenticarte en Compute Engine seleccionando una de las siguientes opciones:
Instala Google Cloud CLI.
Después de la instalación,
inicializa Google Cloud CLI ejecutando el siguiente comando:
Este rol predefinido contiene los permisos necesarios para modificar las recomendaciones de tipo de máquina. Para ver los permisos exactos que son necesarios, expande la sección Permisos requeridos:
Permisos necesarios
Se requieren los siguientes permisos para modificar las recomendaciones de tipo de máquina:
Consulta la configuración de la recomendación de tipo de máquina:
recommender.computeInstanceMachineTypeRecommenderConfig.get
en el proyecto
Modifica la configuración de la recomendación del tipo de máquina:
recommender.computeInstanceMachineTypeRecommenderConfig.update
en el proyecto
La configuración te permite personalizar los siguientes parámetros:
preferred_machine_types: Son las recomendaciones de tipos de máquinas que recibes.
memory_metrics: Cómo se usan las métricas de memoria para generar recomendaciones
Opciones para las recomendaciones de tipos de máquinas
Puedes especificar qué recomendaciones de tipos de máquinas recibes cambiando el parámetro de configuración preferred_machine_types a uno de los siguientes valores:
DEFAULT: El valor predeterminado es STANDARD_AND_CUSTOM.
STANDARD_AND_CUSTOM: Recomienda la mejor coincidencia entre los tipos de máquinas estándar y personalizados.
STANDARD_ONLY: Recomienda la mejor coincidencia solo entre los tipos de máquinas estándar.
Opciones para las recomendaciones de memoria
Para las métricas de memoria que se usan para generar las recomendaciones, puedes establecer los siguientes valores:
MEMORY_METRICS_DEFAULT: Usa las métricas predeterminadas para generar recomendaciones.
MEMORY_METRICS_ALL: Usa todas las métricas disponibles para generar recomendaciones.
MONITORING_AGENT_ONLY: Usa solo las métricas recopiladas por el agente de operaciones de Cloud Monitoring.
Si no usas el agente de operaciones y prefieres no recibir recomendaciones de memoria, usa el valor MEMORY_AGENT_ONLY para inhabilitar las recomendaciones.
Para obtener recomendaciones más precisas sobre el uso de la memoria, usa el Agente de operaciones.
Actualiza la configuración de un proyecto
Para actualizar la configuración de las recomendaciones de tipo de máquina, modifica los valores de configuración y sube la configuración nueva del proyecto.
Puedes cambiar los tipos de máquinas preferidos, las métricas de memoria que se usan para las recomendaciones o ambos valores en la misma solicitud.
Después de subir la configuración nueva, lleva alrededor de 48 horas para que se aplique. Hasta ese momento, las recomendaciones generadas usan la configuración anterior.
ZONE: La zona para la que deseas obtener la configuración actual
LOCAL_FILE_PATH_AND_NAME: Es la ruta de acceso al archivo JSON con la configuración actual.
ETAG: El valor de ETag de la configuración actual.
Puedes encontrar este valor en el resultado del comando anterior.
En el archivo de configuración, especifica solo el objeto recommenderGenerationConfig con los parámetros de configuración actualizados. Si guardaste la configuración en un archivo local, asegúrate de copiar el valor de etag y, luego, quita todos los campos, excepto el valor de recommenderGenerationConfig.
Por ejemplo, para obtener recomendaciones solo con tipos de máquinas estándar, usa el siguiente archivo de configuración:
Si el comando se ejecuta de forma correcta, muestra un cuerpo de respuesta similar al siguiente, que muestra la configuración nueva con los parámetros que acabas de cambiar.
[[["Fácil de comprender","easyToUnderstand","thumb-up"],["Resolvió mi problema","solvedMyProblem","thumb-up"],["Otro","otherUp","thumb-up"]],[["Difícil de entender","hardToUnderstand","thumb-down"],["Información o código de muestra incorrectos","incorrectInformationOrSampleCode","thumb-down"],["Faltan la información o los ejemplos que necesito","missingTheInformationSamplesINeed","thumb-down"],["Problema de traducción","translationIssue","thumb-down"],["Otro","otherDown","thumb-down"]],["Última actualización: 2025-09-09 (UTC)"],[[["\u003cp\u003eThis page guides users on how to customize machine type recommendations at the project level, allowing them to control which machine series are suggested and how memory metrics are used in generating these recommendations.\u003c/p\u003e\n"],["\u003cp\u003eTo modify machine type recommendations, users need the "Compute Recommender Admin" IAM role, which grants the necessary permissions to view and update the recommendation configuration.\u003c/p\u003e\n"],["\u003cp\u003eThe configuration of machine type recommendations involves three main steps: retrieving the current configuration in JSON format, editing this configuration to specify preferred machine types and memory metrics, and uploading the updated file.\u003c/p\u003e\n"],["\u003cp\u003eUsers can specify their preferences for machine type recommendations, choosing between standard and custom types, or standard types only, and can also define how memory metrics are utilized, even opting to rely solely on the Cloud Monitoring Ops Agent or disabling memory recommendations.\u003c/p\u003e\n"],["\u003cp\u003eAfter uploading a new configuration, it takes approximately 48 hours for the changes to take effect, and during this time, the system will continue to use the previous settings for generating any recommendations.\u003c/p\u003e\n"]]],[],null,["# Configure machine type recommendations\n\n*** ** * ** ***\n\n|\n| **Preview**\n|\n|\n| This product or feature is subject to the \"Pre-GA Offerings Terms\" in the General Service Terms section\n| of the [Service Specific Terms](/terms/service-terms#1).\n|\n| Pre-GA products and features are available \"as is\" and might have limited support.\n|\n| For more information, see the\n| [launch stage descriptions](/products#product-launch-stages).\n\nThis page describes how to view and modify the configuration of your machine\ntype recommendations. You can modify which machine types are recommended,\nso the generated recommendations only include your preferred machine series.\nYou can also change the metrics used to generate memory recommendations to\nimprove the accuracy of the recommendations. The changes to the configuration\nare applied at the project level.\n\nFor more information about machine type recommendations, see\n[Apply machine type recommendations for VMs](/compute/docs/instances/apply-machine-type-recommendations-for-instances).\n\nBefore you begin\n----------------\n\n- If you haven't already, set up [authentication](/compute/docs/authentication). Authentication verifies your identity for access to Google Cloud services and APIs. To run code or samples from a local development environment, you can authenticate to Compute Engine by selecting one of the following options:\n 1.\n [Install](/sdk/docs/install) the Google Cloud CLI.\n\n After installation,\n [initialize](/sdk/docs/initializing) the Google Cloud CLI by running the following command:\n\n ```bash\n gcloud init\n ```\n\n\n If you're using an external identity provider (IdP), you must first\n [sign in to the gcloud CLI with your federated identity](/iam/docs/workforce-log-in-gcloud).\n | **Note:** If you installed the gcloud CLI previously, make sure you have the latest version by running `gcloud components update`.\n 2. [Set a default region and zone](/compute/docs/gcloud-compute#set_default_zone_and_region_in_your_local_client).\n\n### Required roles\n\n\nTo get the permissions that\nyou need to modify machine type recommendations,\n\nask your administrator to grant you the\n\n\n[Compute Recommender Admin](/iam/docs/roles-permissions/recommender#recommender.computeAdmin) (`roles/recommender.computeAdmin`)\nIAM role on your project.\n\n\nFor more information about granting roles, see [Manage access to projects, folders, and organizations](/iam/docs/granting-changing-revoking-access).\n\n\nThis predefined role contains\n\nthe permissions required to modify machine type recommendations. To see the exact permissions that are\nrequired, expand the **Required permissions** section:\n\n\n#### Required permissions\n\nThe following permissions are required to modify machine type recommendations:\n\n- View the machine type recommendation configuration: `recommender.computeInstanceMachineTypeRecommenderConfig.get` on the project\n- Modify the machine type recommendation configuration: `recommender.computeInstanceMachineTypeRecommenderConfig.update` on the project\n\n\nYou might also be able to get\nthese permissions\nwith [custom roles](/iam/docs/creating-custom-roles) or\nother [predefined roles](/iam/docs/roles-overview#predefined).\n\nConfigure your recommendations\n------------------------------\n\nTo configure your machine type recommendations, perform the following steps:\n\n1. [Get the current configuration](#get_current_configuration) to obtain the JSON file of the configuration.\n2. Edit the configuration file in JSON format.\n3. [Upload the new configuration file](#update_configuration) to apply the changes.\n\n### Get the current configuration\n\nYou can obtain the current configuration for machine type recommendations for\nyour project by using the Google Cloud CLI. \n\n### gcloud\n\nTo get the current configuration parameters, use the\n[`gcloud beta recommender recommender-config describe` command](/sdk/gcloud/reference/beta/recommender/recommender-config/describe)\nand use `google.compute.instance.MachineTypeRecommender` as the\nRecommender ID: \n\n```\ngcloud beta recommender recommender-config describe google.compute.instance.MachineTypeRecommender \\\n --project=PROJECT_ID \\\n --location=ZONE\n```\n\nReplace the following:\n\n- \u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e: the ID of your project.\n- \u003cvar translate=\"no\"\u003eZONE\u003c/var\u003e: the zone for which you want to obtain the current configuration.\n\nThe command returns the recommender configuration, including the modifiable\nsubfields `preferred_machine_types` and `memory_metrics`. \n\n```\n{\n \"name\": \"projects/PROJECT_ID/locations/ZONE/recommenders/google.compute.instance.MachineTypeRecommender/config\",\n \"recommenderGenerationConfig\": {\n \"params\": {\n \"preferred_machine_types\": \"STANDARD_AND_CUSTOM\",\n \"memory_metrics\": \"MONITORING_AGENT_ONLY\"\n }\n },\n \"etag\": \"\\\"2f3c9b65cda6a4ba\\\"\",\n \"updateTime\": \"2024-08-20T04:41:15.330351Z\"\n}\n```\n\nIf you want to save the configuration to a local file, so you can edit it\nand upload it later, run the following command: \n\n```\ngcloud beta recommender recommender-config describe \\\n google.compute.instance.MachineTypeRecommender \\\n --project=PROJECT_ID \\\n --location=ZONE \u003e LOCAL_FILE_PATH_AND_NAME\n```\n\n### Edit the recommendation configuration\n\nThe configuration lets you customize the following parameters:\n\n- `preferred_machine_types`: The machine types recommendations that you receive.\n- `memory_metrics`: How memory metrics are used to generate recommendations.\n\n#### Options for machine types recommendations\n\nYou can specify which machine types recommendations you get by changing\nthe `preferred_machine_types` setting to one of the following values:\n\n- `DEFAULT`: The default value is `STANDARD_AND_CUSTOM`.\n- `STANDARD_AND_CUSTOM`: Recommends the best match from standard and custom machine types.\n- `STANDARD_ONLY`: Recommends the best match from standard machine types only.\n\n#### Options for memory recommendations\n\nFor the memory metrics used to generate the recommendations, you can set\nthe following values:\n\n- `MEMORY_METRICS_DEFAULT`: Use the default metrics to generate recommendations.\n- `MEMORY_METRICS_ALL`: Use all the available metrics to generate recommendations.\n- `MONITORING_AGENT_ONLY`: Use only metrics collected by the Cloud Monitoring Ops Agent.\n- If you don't use the Ops Agent, and prefer not to receive memory recommendations, use the `MEMORY_AGENT_ONLY` value to disable the recommendations.\n\nTo get more accurate recommendations about memory usage, use the\n[Ops Agent](/monitoring/agent/ops-agent).\n\n### Update the configuration for a project\n\nTo update the configuration of your machine type recommendations, modify the\nconfiguration values and upload the new configuration for your project.\nYou can change the preferred machine types, the memory metrics used for\nrecommendations, or both values in the same request.\n\nAfter you upload the new configuration, it takes approximately 48 hours for it\nto take effect. Until that time, any generated recommendations use the previous\nconfiguration. \n\n### gcloud\n\nTo update the configuration parameters, use the\n[`gcloud beta recommender recommender-config update` command](/sdk/gcloud/reference/beta/recommender/recommender-config/update)\nand use `google.compute.instance.MachineTypeRecommender` as the\nRecommender ID: \n\n```\ngcloud beta recommender recommender-config update google.compute.instance.MachineTypeRecommender\\\n --project=PROJECT_ID \\\n --location=ZONE \\\n --config-file=LOCAL_FILE_PATH_AND_NAME \\\n --etag=ETAG\n```\n\nReplace the following:\n\n- \u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e: the ID of your project.\n- \u003cvar translate=\"no\"\u003eZONE\u003c/var\u003e: the zone for which you want to obtain the current configuration.\n- \u003cvar translate=\"no\"\u003eLOCAL_FILE_PATH_AND_NAME\u003c/var\u003e: the path to the JSON file with the current configuration.\n- \u003cvar translate=\"no\"\u003eETAG\u003c/var\u003e: the etag value of the current configuration. You can find this value in the output of the [previous command](#get_current_configuration).\n\nIn your configuration file, specify only the `recommenderGenerationConfig`\nobject with the updated configuration parameters. If you saved the\nconfiguration to a local file, make sure you copy the `etag` value, then\nremove all the fields except the `recommenderGenerationConfig` value.\n\nFor example, to get recommendations with standard machine types only,\nuse the following configuration file: \n\n```\n{\n \"params\": {\n \"preferred_machine_types\": \"STANDARD_ONLY\"\n }\n}\n```\n\nIf the command is successful, it returns a response body similar to the\nfollowing, showing the new configuration with the parameters you just\nmodified. \n\n```\n{\n \"name\": \"projects/PROJECT_ID/locations/ZONE/recommenders/google.compute.instance.MachineTypeRecommender/config\",\n \"recommenderGenerationConfig\": {\n \"params\": {\n \"preferred_machine_types\": \"STANDARD_ONLY\"\n }\n },\n \"etag\": \"\\\"1245f9435cd01ea8\\\"\",\n \"updateTime\": \"2024-08-20T05:00:05Z\"\n}\n```\n| **Note:** the `etag` value changes from the previous configuration. You must use the new value if you update the configuration again.\n\nWhat's next\n-----------\n\n- Learn how to [Apply machine type recommendations to instances](/compute/docs/instances/apply-machine-type-recommendations-for-instances).\n- Learn how to [Apply machine type recommendations to managed instance groups](/compute/docs/instance-groups/apply-machine-type-recommendations-managed-instance-groups)."]]
