AWS Certified Data Engineer Associate DEA-C01 Practice Question

A data engineering team ingests 10 TB of semi-structured clickstream events through Amazon Kinesis Data Firehose into an Amazon S3 data lake. Analysts will run frequent Amazon Athena queries that usually reference only a few columns. The team wants to minimize both query latency and the amount of data scanned to control cost. Which file format should the team configure Kinesis Data Firehose to deliver to S3 to best meet these requirements?

Apache Parquet files with Snappy compression
Gzip-compressed JSON files
Line-delimited JSON text (.txt)
Comma-separated values (CSV)

Report Issue

Answer Description

Amazon Athena charges by the amount of data it reads from Amazon S3. Columnar, splittable formats such as Apache Parquet store each column contiguously and include column statistics. When Athena reads only a subset of columns, far less data is scanned, which lowers query cost and improves performance. Row-oriented formats like CSV, raw JSON, or even compressed JSON still require Athena to read the full row set and do not provide predicate pushdown. Therefore, storing the data in Parquet with a compression codec such as Snappy is the most cost-effective and performant choice for the described workload.

Ask Bash

Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.