A gaming company collects clickstream events with Amazon Kinesis Data Streams and uses Kinesis Data Firehose to deliver JSON files to Amazon S3. Data scientists query the most recent data in Amazon Athena. The company wants to reduce Athena query costs without rebuilding the existing data pipeline. Which option minimizes management effort while meeting the requirement?
Choose an answer
Tap an option to check your answer.
Correct answer: Configure Firehose to output Apache Parquet, set a custom S3 object prefix like YYYYMMDD and a large buffer size. For historical data, run an AWS Glue ETL job to coalesce small JSON files into larger Parquet files with the YYYYMMDD prefix, then run ALTER TABLE ADD PARTITION to update the existing Athena table..
Why this is the answer
The correct option minimizes Athena query costs by converting data to Parquet, which is a columnar format optimized for analytical queries, and by creating larger files (via Firehose buffer size and Glue ETL) to reduce S3 object overhead. Using a YYYYMMDD prefix enables S3 partitioning, further reducing data scanned by Athena. This approach leverages existing services (Firehose, S3, Athena, Glue) with minimal new infrastructure. Incorrect options: Creating a Spark job on EMR adds operational overhead for managing EMR clusters. Using Amazon Managed Service for Apache Flink introduces a new, more complex service and additional cost. Attaching a Lambda to Firehose for conversion is inefficient for high-volume streaming data due to Lambda's invocation model and potential for creating many small files.
Pass your exam — without the endless answer hunt
Get every verified question and explanation for this exam in one place, and save hours of prep. 1,000+ certifications · 20+ languages · free to start.
Pass your exam faster → No card needed