Executar um pipeline em um cluster atual do Dataproc
Mantenha tudo organizado com as coleções
Salve e categorize o conteúdo com base nas suas preferências.
Nesta página, descrevemos como executar um pipeline no Cloud Data Fusion em um cluster atual do Dataproc.
Por padrão, o Cloud Data Fusion cria clusters efêmeros para cada pipeline: cria um cluster no início da execução do pipeline e o exclui após a conclusão da execução do pipeline. Embora esse comportamento economize custos, garantindo que os recursos sejam criados somente quando necessário, esse comportamento padrão pode não ser desejável nos seguintes cenários:
Se o tempo necessário para criar um novo cluster para cada pipeline for adequado para seu caso de uso.
Se a organização exigir que a criação de cluster seja gerenciada centralmente; por exemplo, quando você quiser aplicar políticas específicas a todos os clusters do Dataproc.
Para esses cenários, execute pipelines em um cluster atual seguindo as etapas a seguir.
Nas versões 6.2.1 e posteriores do Cloud Data Fusion, é possível se conectar a um cluster atual do Dataproc quando você cria um novo perfil do Compute Engine.
Acesse sua instância:
No console Google Cloud , acesse a página do Cloud Data Fusion.
Para abrir a instância no Cloud Data Fusion Studio, clique em Instâncias e em Ver instância.
[[["Fácil de entender","easyToUnderstand","thumb-up"],["Meu problema foi resolvido","solvedMyProblem","thumb-up"],["Outro","otherUp","thumb-up"]],[["Difícil de entender","hardToUnderstand","thumb-down"],["Informações incorretas ou exemplo de código","incorrectInformationOrSampleCode","thumb-down"],["Não contém as informações/amostras de que eu preciso","missingTheInformationSamplesINeed","thumb-down"],["Problema na tradução","translationIssue","thumb-down"],["Outro","otherDown","thumb-down"]],["Última atualização 2025-08-12 UTC."],[[["\u003cp\u003eThis guide explains how to run Cloud Data Fusion pipelines against a pre-existing Dataproc cluster, instead of the default behavior of creating and deleting ephemeral clusters.\u003c/p\u003e\n"],["\u003cp\u003eUsing an existing cluster can be beneficial when cluster creation time is prohibitive or when centralized cluster management is required by the organization.\u003c/p\u003e\n"],["\u003cp\u003eTo use an existing Dataproc cluster, a Cloud Data Fusion instance and a pre-created Dataproc cluster are needed, and if running version 6.2 of Cloud Data Fusion, an older Dataproc image or an upgrade is required.\u003c/p\u003e\n"],["\u003cp\u003eConnecting to the existing cluster involves creating a new Compute Engine profile within Cloud Data Fusion and selecting the "Existing Dataproc" option, then providing the required information.\u003c/p\u003e\n"],["\u003cp\u003eAfter creating the custom profile, the pipeline must be configured in the Studio to use the custom profile, and then the pipeline will run against the designated Dataproc cluster.\u003c/p\u003e\n"]]],[],null,["# Run a pipeline against an existing Dataproc cluster\n\nThis page describes how to run a pipeline in Cloud Data Fusion against\nan existing Dataproc cluster.\n\nBy default, Cloud Data Fusion creates ephemeral clusters for each pipeline:\nit creates a cluster at the beginning of the pipeline run, and then deletes it\nafter the pipeline run completes. While this behavior saves costs by ensuring\nthat resources are only created when required, this default behavior might not\nbe desirable in the following scenarios:\n\n- If the time it takes to create a new cluster for every pipeline is\n prohibitive for your use case.\n\n- If your organization requires cluster creation to be managed centrally; for\n example, when you want to enforce certain policies for all\n Dataproc clusters.\n\nFor these scenarios, you instead run pipelines against an existing cluster with\nthe following steps.\n\nBefore you begin\n----------------\n\nYou need the following:\n\n- A Cloud Data Fusion instance.\n\n [Create a Cloud Data Fusion instance](/data-fusion/docs/how-to/create-instance)\n- An existing Dataproc cluster.\n\n [Create a Dataproc cluster](/dataproc/docs/guides/create-cluster)\n- If you run your pipelines in Cloud Data Fusion version 6.2, use an\n older [Dataproc image](/dataproc/docs/concepts/versioning/dataproc-versions)\n that runs with Hadoop 2.x (for example, 1.5-debian10), or [upgrade to the\n latest Cloud Data Fusion version](/data-fusion/docs/how-to/upgrading).\n\nConnect to the existing cluster\n-------------------------------\n\nIn Cloud Data Fusion versions 6.2.1 and later, you can connect to an\nexisting Dataproc cluster when you create a new Compute Engine\nprofile.\n\n1. Go to your instance:\n\n\n 1. In the Google Cloud console, go to the Cloud Data Fusion page.\n\n 2. To open the instance in the Cloud Data Fusion Studio,\n click **Instances** , and then click **View instance**.\n\n [Go to Instances](https://console.cloud.google.com/data-fusion/locations/-/instances)\n\n \u003cbr /\u003e\n\n2. Click **System admin**.\n\n3. Click the **Configuration** tab.\n\n4. Click\n expand_more\n **System compute profiles**.\n\n5. Click **Create new profile**. A page of provisioners opens.\n\n6. Click **Existing Dataproc**.\n\n7. Enter the profile, cluster, and monitoring information.\n\n8. Click **Create**.\n\nConfigure your pipeline to use the custom profile\n-------------------------------------------------\n\n1. Go to your instance:\n\n\n 1. In the Google Cloud console, go to the Cloud Data Fusion page.\n\n 2. To open the instance in the Cloud Data Fusion Studio,\n click **Instances** , and then click **View instance**.\n\n [Go to Instances](https://console.cloud.google.com/data-fusion/locations/-/instances)\n\n \u003cbr /\u003e\n\n2. Go to your pipeline on the **Studio** page.\n\n3. Click **Configure**.\n\n4. Click **Compute config**.\n\n5. Click the profile that you created.\n\n **Figure 1**: Click the custom profile\n6. Run the pipeline. It runs against the existing Dataproc\n cluster.\n\nWhat's next\n-----------\n\n- Learn more about [configuring clusters](/data-fusion/docs/concepts/configure-clusters).\n- Troubleshoot [deleting clusters](/data-fusion/docs/troubleshoot-deleting-clusters)."]]
