Google Professional Machine Learning Engineer Question 391
Single answerGoogle Cloud PlatformYou are tasked with building a machine learning workflow to preprocess data, train a model, and deploy it for online predictions. Your organization uses Google Cloud, and the workflow must be automated, reusable, and scalable. Which orchestration framework should you use to achieve this, while also integrating well with other Google Cloud services?
- A
Kubeflow Pipelines
- B
Apache Airflow
- C
Vertex AI Pipelines
- D
Cloud Function Workflows
Show answer and explanation
Correct answer: C
Explanation
Vertex AI Pipelines is the optimal choice for orchestrating machine learning workflows in Google Cloud. It is a fully managed service that provides tight integration with Google Cloud's machine learning and data processing services, enabling users to build reusable, scalable, and automated workflows with minimal effort.
- A. Incorrect.
Kubeflow Pipelines is a powerful tool for orchestrating machine learning workflows, but it is less integrated with Google Cloud's managed services like Vertex AI. It would require more manual effort to connect with other services.
- B. Incorrect.
Apache Airflow is a general-purpose orchestration tool that can be used for ML workflows, but it is not optimized for machine learning-specific tasks or Google Cloud's ML ecosystem.
- C. Correct.
Vertex AI Pipelines is the best choice as it is specifically designed for machine learning workflows and provides seamless integration with other Google Cloud services like BigQuery, Cloud Storage, and Vertex AI for training and deployment.
- D. Incorrect.
Cloud Function Workflows is a general-purpose workflow automation solution in Google Cloud, but it is not specifically tailored for machine learning pipelines.