Athena¶

• serverless
• uses SQL Language
• interactive query serive to analyse data in s3
• per only per queries run
• process logs
• ad hoc analysis
• expensive if data is not columnar
• store results back to S3

Use¶

• commonly used with quicksight
• buisness intel, alaytics, reporting
• use glue to covert data to parquet or orc
• partition data by path

pathtoBucket/pathtotable
                        /partition_col_key
                        /partionen_col_key

• use larger files

Supported Data types¶

• CSV
• JSON
• ORC
• Avro
• Parquet (faster than csv)

Price¶

• 5 Dollar per TB scanned

Federated query¶

• query data from outside from S3
• use data source connectior (Lambda function which connects to another service)