A gaming company is collecting clickstream data into multiple Amazon Kinesis data streams. The company uses Amazon Kinesis Data Firehose delivery streams to store the data in JSON format in Amazon S3. Data scientists use Amazon Athena to query the most recent data and derive business insights. The company wants to reduce its Athena costs without having to recreate the data pipeline. The company prefers a solution that will require less management effort.
Which set of actions can the data scientists take immediately to reduce costs?
A. Change the Kinesis Data Firehose output format to Apache Parquet. Provide a custom S3 object YYYYMMDD prefix expression and specify a large buffer size. For the existing data, run an AWS Glue ETL job to combine and convert small JSON files to large Parquet files and add the YYYYMMDD prefix. Use ALTER TABLE ADD PARTITION to reflect the partition on the existing Athena table.
B. Create an Apache Spark job that combines and converts JSON files to Apache Parquet files. Launch an Amazon EMR ephemeral cluster daily to run the Spark job to create new Parquet files in a different S3 location. Use ALTER TABLE SET LOCATION to reflect the new S3 location on the existing Athena table.
C. Create a Kinesis data stream as a delivery target for Kinesis Data Firehose. Run Apache Flink on Amazon Kinesis Data Analytics on the stream to read the streaming data, aggregate it. and save it to Amazon S3 in Apache Parquet format with a custom S3 object YYYYMMDD prefix. Use ALTER TABLE ADD PARTITION to reflect the partition on the existing Athena table.
D. Integrate an AWS Lambda function with Kinesis Data Firehose to convert source records to Apache Parquet and write them to Amazon S3. In parallel, run an AWS Glue ETL job to combine and convert existing JSON files to large Parquet files. Create a custom S3 object YYYYMMDD prefix. Use ALTER TABLE ADD PARTITION to reflect the partition on the existing Athena table.
Which set of actions can the data scientists take immediately to reduce costs?
A. Change the Kinesis Data Firehose output format to Apache Parquet. Provide a custom S3 object YYYYMMDD prefix expression and specify a large buffer size. For the existing data, run an AWS Glue ETL job to combine and convert small JSON files to large Parquet files and add the YYYYMMDD prefix. Use ALTER TABLE ADD PARTITION to reflect the partition on the existing Athena table.
B. Create an Apache Spark job that combines and converts JSON files to Apache Parquet files. Launch an Amazon EMR ephemeral cluster daily to run the Spark job to create new Parquet files in a different S3 location. Use ALTER TABLE SET LOCATION to reflect the new S3 location on the existing Athena table.
C. Create a Kinesis data stream as a delivery target for Kinesis Data Firehose. Run Apache Flink on Amazon Kinesis Data Analytics on the stream to read the streaming data, aggregate it. and save it to Amazon S3 in Apache Parquet format with a custom S3 object YYYYMMDD prefix. Use ALTER TABLE ADD PARTITION to reflect the partition on the existing Athena table.
D. Integrate an AWS Lambda function with Kinesis Data Firehose to convert source records to Apache Parquet and write them to Amazon S3. In parallel, run an AWS Glue ETL job to combine and convert existing JSON files to large Parquet files. Create a custom S3 object YYYYMMDD prefix. Use ALTER TABLE ADD PARTITION to reflect the partition on the existing Athena table.