AI-102 Question 304
Select 3You are designing an Azure AI solution for a company that needs to preprocess data for a machine learning model. The company stores its data in multiple sources, including an Azure SQL Database, an Azure Blob Storage container, and a third-party on-premises PostgreSQL database. To streamline the data import process during model training, which data sources can you directly import into Azure Machine Learning datasets?
- A
Azure SQL Database
- B
Azure Blob Storage
- C
On-premises PostgreSQL database via a secured connection
- D
Google BigQuery
- E
Azure Cosmos DB
Show answer and explanation
Correct answers: A, B, E
Explanation
Azure Machine Learning supports direct data imports from various Azure-native sources, such as Azure SQL Database, Azure Blob Storage, and Azure Cosmos DB. However, non-Azure sources like on-premises databases or third-party cloud services require additional integration solutions such as Data Factory or custom ETL processes to transfer data into supported formats or storage locations.
- A. Correct.
Azure SQL Database is a supported data source for Azure Machine Learning datasets, allowing direct imports.
- B. Correct.
Azure Blob Storage is a native and highly compatible data source in Azure Machine Learning for importing datasets.
- C. Incorrect.
On-premises PostgreSQL databases cannot be directly imported into Azure Machine Learning datasets, but data can be transferred by other means (e.g., Data Factory).
- D. Incorrect.
Google BigQuery is not natively supported as a direct import source for Azure Machine Learning datasets.
- E. Correct.
Azure Cosmos DB is a supported data source for direct imports into Azure Machine Learning datasets.