[[["容易理解","easyToUnderstand","thumb-up"],["確實解決了我的問題","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["難以理解","hardToUnderstand","thumb-down"],["資訊或程式碼範例有誤","incorrectInformationOrSampleCode","thumb-down"],["缺少我需要的資訊/範例","missingTheInformationSamplesINeed","thumb-down"],["翻譯問題","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["上次更新時間:2025-09-04 (世界標準時間)。"],[[["\u003cp\u003eThis document details how to enable and configure Transformation Pushdown in Cloud Data Fusion to execute pipeline transformations directly within BigQuery instead of Spark.\u003c/p\u003e\n"],["\u003cp\u003eTransformation Pushdown, available in Cloud Data Fusion version 6.5.0 and later, can be enabled on a deployed pipeline through the Cloud Data Fusion Studio by navigating to the 'Transformation Pushdown' section in the pipeline configuration settings.\u003c/p\u003e\n"],["\u003cp\u003eVarious optional configurations are available for Transformation Pushdown, including settings for project ID, service account details, temporary bucket name, encryption keys, and more, which are adjustable depending on the Cloud Data Fusion version.\u003c/p\u003e\n"],["\u003cp\u003ePipeline runtime logs offer insights into the SQL queries executed in BigQuery, enabling users to monitor which stages are being pushed down and track the progress of push, join, and pull operations within the BigQuery environment.\u003c/p\u003e\n"],["\u003cp\u003eThe BigQuery Job ID, which is logged, can be used to view detailed information on resource utilization and potential errors to better track the transformations.\u003c/p\u003e\n"]]],[],null,["# Execute transformations in BigQuery\n\nThis page describes how to execute transformations to BigQuery\ninstead of Spark in Cloud Data Fusion.\n\nFor more information, see the\n[Transformation Pushdown overview](/data-fusion/docs/concepts/transformation-pushdown).\n\nBefore you begin\n----------------\n\nTransformation Pushdown is available in version 6.5.0 and later. If your\npipeline runs in an earlier environment, you can\n[upgrade your instance](/data-fusion/docs/how-to/upgrading) to the latest\nversion.\n\nEnable Transformation Pushdown on your pipeline\n-----------------------------------------------\n\n### Console\n\n\nTo enable Transformation Pushdown on a deployed pipeline, do the\nfollowing:\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 menu **Menu \\\u003e**\n list **List**.\n\n The deployed pipeline tab opens.\n3. Click the desired deployed pipeline to open it in the **Pipeline\n Studio**.\n\n4. Click **Configure \\\u003e Transformation Pushdown**.\n\n5. Click **Enable Transformation Pushdown**.\n\n6. In the **Dataset** field, enter a BigQuery dataset name.\n\n Optional: To use a macro, click **M** . For more information, see\n [Datasets](/bigquery/docs/datasets-intro#datasets).\n7. Optional: Configure the options, if needed.\n\n8. Click **Save**.\n\n### Optional configurations\n\nMonitor performance changes in the logs\n---------------------------------------\n\nThe pipeline runtime logs include messages that show the SQL queries that are\nrun in BigQuery. You can monitor which stages in the pipeline get\npushed into BigQuery.\n\nThe following example shows the log entries when pipeline execution begins. The\nlogs indicate that the `JOIN` operations in your pipeline have been pushed down\nBigQuery for execution: \n\n INFO [Driver:i.c.p.g.b.s.BigQuerySQLEngine@190] - Validating join for stage 'Users' can be executed on BigQuery: true\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@131] - Starting push for dataset 'UserProfile'\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@131] - Starting push for dataset 'UserDetails'\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@292] - Starting join for dataset 'Users'\n INFO [Driver:i.c.p.g.b.s.BigQuerySQLEngine@190] - Validating join for stage 'UserPurchases' can be executed on BigQuery: true\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@131] - Starting push for dataset 'Purchases'\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@292] - Starting join for dataset 'UserPurchases'\n INFO [Driver:i.c.p.g.b.s.BigQuerySQLEngine@190] - Validating join for stage 'MostPopularNames' can be executed on BigQuery: true\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@131] - Starting push for dataset 'FirstNameCounts'\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@292] - Starting join for dataset 'MostPopularNames'\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@193] - Starting pull for dataset 'MostPopularNames'\n\nThe following example shows the table names that will be assigned for each of\nthe datasets involved in the pushdown execution: \n\n INFO [batch-sql-engine-adapter:i.c.p.g.b.s.BigQuerySQLEngine@145] - Executing Push operation for dataset Purchases stored in table \u003cTABLE_ID\u003e\n INFO [batch-sql-engine-adapter:i.c.p.g.b.s.BigQuerySQLEngine@145] - Executing Push operation for dataset UserDetails stored in table \u003cTABLE_ID\u003e\n INFO [batch-sql-engine-adapter:i.c.p.g.b.s.BigQuerySQLEngine@145] - Executing Push operation for dataset FirstNameCounts stored in table \u003cTABLE_ID\u003e\n INFO [batch-sql-engine-adapter:i.c.p.g.b.s.BigQuerySQLEngine@145] - Executing Push operation for dataset UserProfile stored in table \u003cTABLE_ID\u003e\n\nAs the execution continues, the logs show the completion of push stages, and\neventually the execution of `JOIN` operations. For example: \n\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@133] - Completed push for dataset 'UserProfile'\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@133] - Completed push for dataset 'UserDetails'\n DEBUG [batch-sql-engine-adapter:i.c.p.g.b.s.BigQuerySQLEngine@235] - Executing join operation for dataset Users\n INFO [batch-sql-engine-adapter:i.c.p.g.b.s.BigQueryJoinDataset@118] - Creating table `\u003cTABLE_ID\u003e` using job: \u003cJOB_ID\u003e with SQL statement: SELECT `UserDetails`.id AS `id` , `UserDetails`.first_name AS `first_name` , `UserDetails`.last_name AS `last_name` , `UserDetails`.email AS `email` , `UserProfile`.phone AS `phone` , `UserProfile`.profession AS `profession` , `UserProfile`.age AS `age` , `UserProfile`.address AS `address` , `UserProfile`.score AS `score` FROM `your_project.your_dataset.\u003cDATASET_ID\u003e` AS `UserProfile` LEFT JOIN `your_project.your_dataset.\u003cDATASET_ID\u003e` AS `UserDetails` ON `UserProfile`.id = `UserDetails`.id\n INFO [batch-sql-engine-adapter:i.c.p.g.b.s.BigQueryJoinDataset@151] - Created BigQuery table `\u003cTABLE_ID\u003e\n INFO [batch-sql-engine-adapter:i.c.p.g.b.s.BigQuerySQLEngine@245] - Executed join operation for dataset Users\n\nWhen all stages have completed, a message shows that the `Pull` operation has\nbeen completed. This indicates that the BigQuery export process\nhas been triggered and records will start being read into the pipeline after\nthis export job begins. For example: \n\n DEBUG [batch-sql-engine-adapter:i.c.c.e.s.b.BatchSQLEngineAdapter@196] - Completed pull for dataset 'MostPopularNames'\n\nIf the pipeline execution encounters errors, they are described in the logs.\n\nFor details about the execution of the BigQuery `JOIN`\noperations, such as resource utilization, execution time, and error causes, you\ncan view the BigQuery Job data using the Job ID, which appears in\nthe job logs.\n\nReview pipeline metrics\n-----------------------\n\nFor more information about the metrics that Cloud Data Fusion provides for\nthe part of the pipeline that's executed in BigQuery, see\n[BigQuery pushdown pipeline metrics](https://cdap.atlassian.net/wiki/spaces/DOCS/pages/1822392321/BigQuery+Pushdown+Pipeline+Metrics).\n\nWhat's next\n-----------\n\n- Learn more about [Transformation Pushdown](/data-fusion/docs/concepts/transformation-pushdown) in Cloud Data Fusion."]]