QuestionQ16

VerifiedData Ingestion and Transformation

A company is creating a data lake for a new analytics team. The company uses Amazon S3 for storage and Amazon Athena for query analysis. All data stored in Amazon S3 uses Apache Parquet format.

The company is operating a new Oracle database as a source system in its data center. The Oracle database contains 70 tables, and every table has a primary key. Data in the source system can change occasionally. The company wants to ingest the tables into the data lake each day.

Which solution meets this requirement with the LEAST effort?

Explanation

AWS DMS can perform ongoing change data capture (CDC) from an Oracle source and write both full-load and change data to Amazon S3. Its S3 target supports Apache Parquet output, so it handles changed source records without building custom incremental logic or triggers. AWS Glue JDBC bookmarks track previously read rows by bookmark keys and do not capture updates to existing rows.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!