nerdexam
Google

PROFESSIONAL-MACHINE-LEARNING-ENGINEER · Question #200

You are implementing a batch inference ML pipeline in Google Cloud. The model was developed using TensorFlow and is stored in SavedModel format in Cloud Storage. You need to apply the model to a…

The correct answer is D. Configure a Vertex AI batch prediction job to apply the model to the historical data in BigQuery. To perform batch inference with a TensorFlow SavedModel on a 10 TB BigQuery dataset, configure a Vertex AI batch prediction job to directly read data from the BigQuery table.

Submitted by kevin_r· Apr 18, 2026ML pipeline operationalization

Question

You are implementing a batch inference ML pipeline in Google Cloud. The model was developed using TensorFlow and is stored in SavedModel format in Cloud Storage. You need to apply the model to a historical dataset containing 10 TB of data that is stored in a BigQuery table. How should you perform the inference?

Options

  • AExport the historical data to Cloud Storage in Avro format. Configure a Vertex AI batch prediction
  • BImport the TensorFlow model by using the CREATE MODEL statement in BigQuery ML. Apply
  • CExport the historical data to Cloud Storage in CSV format. Configure a Vertex AI batch prediction
  • DConfigure a Vertex AI batch prediction job to apply the model to the historical data in BigQuery

How the community answered

(22 responses)
  • A
    9% (2)
  • B
    5% (1)
  • C
    5% (1)
  • D
    82% (18)

Why each option

To perform batch inference with a TensorFlow SavedModel on a 10 TB BigQuery dataset, configure a Vertex AI batch prediction job to directly read data from the BigQuery table.

AExport the historical data to Cloud Storage in Avro format. Configure a Vertex AI batch prediction

Exporting 10 TB of historical data from BigQuery to Cloud Storage adds an unnecessary and potentially costly data transfer and storage step, as Vertex AI can read directly from BigQuery.

BImport the TensorFlow model by using the CREATE MODEL statement in BigQuery ML. Apply

While BigQuery ML allows importing TensorFlow models, the question specifies applying an existing SavedModel to historical data using a batch inference pipeline, which Vertex AI batch prediction is explicitly designed for.

CExport the historical data to Cloud Storage in CSV format. Configure a Vertex AI batch prediction

Exporting 10 TB of historical data from BigQuery to Cloud Storage adds an unnecessary and potentially costly data transfer and storage step, as Vertex AI can read directly from BigQuery.

DConfigure a Vertex AI batch prediction job to apply the model to the historical data in BigQueryCorrect

Vertex AI batch prediction jobs can directly read input data from BigQuery tables, which is efficient for large datasets. This approach eliminates the need for intermediate data export to Cloud Storage, simplifying the pipeline and reducing costs.

Concept tested: Vertex AI Batch Prediction with BigQuery input

Source: https://cloud.google.com/vertex-ai/docs/predictions/batch-predictions#bigquery_input_output

Topics

#Vertex AI#Batch Prediction#BigQuery#ML Pipeline

Community Discussion

No community discussion yet for this question.

Sign up to join the discussion
Full PROFESSIONAL-MACHINE-LEARNING-ENGINEER Practice