DEA-C01 Practice Question 6: DynamoDB Export S3 Analytics
AWS Certified Data Engineer – Associate (DEA-C01) · Test 1 · question 6 of 10
Analysts need to run monthly SQL analysis over the full contents of a 2 TB Amazon DynamoDB orders table. The table serves a latency-sensitive checkout API in on-demand capacity mode, and the operations team requires that the analysis have no effect on the table's read performance or throughput cost. The analysts already use Amazon Athena for everything else in the data lake, and the data engineering team wants to write as little code as possible.
Which solution should the data engineer choose?
Show answer and explanation
Answer:
DynamoDB export to S3 is purpose-built for this. It reads from the table's continuous backups, so it requires point-in-time recovery to be enabled, and the documentation states plainly that exports "are asynchronous, they don't consume read capacity units (RCUs) and have no impact on table performance and availability". Output lands in Amazon S3 as DynamoDB JSON or Amazon Ion, explicitly so that you can "perform analytics and complex queries on your data using other AWS services such as Athena, AWS Glue, Amazon SageMaker AI, Amazon EMR, and AWS Lake Formation". It is a console or API call, no pipeline code at all, and both full and incremental exports are supported.
Option A reads through the live table. A parallel Scan of 2 TB consumes real read capacity, competes with checkout traffic for partition-level throughput, and in on-demand mode is billed as read requests. That is exactly the impact the operations team ruled out.
Option B captures only *future* changes. DynamoDB Streams has a 24-hour retention window and emits change records, so it can never reproduce the 2 TB of history already in the table. It is also the most code of any option, and it makes the pipeline responsible for compaction and deduplication.
Option D also hits the live table. The Athena DynamoDB federated connector issues Scans against the real table at query time, so a monthly full-table analytical query would consume capacity and add latency to the checkout path.