Serverless Data Analysis and Reporting using SQL
summary
Demonstrates running SQL against CSV on S3 via Amazon Athena, creating lightweight table schemas (schema‑on‑read), and a demo join that reads many files without loading data first. Explains that Athena wraps Trino backend, returns asynchronous query IDs so clients must poll and retry, defaults time out at 30 minutes, and charges about $5 per TB scanned. Notes support for column formats (Parquet/ORC), and surprising ACID via Iceberg for transactional tables.
AI-written summary · about
chapters (17)
- 0:00Solution Street introduction
- 0:44Why Amazon Athena
- 3:18Athena features and benefits
- 4:21Live Athena query demo
- 7:33Complex SQL use cases
- 8:17Typical Athena use cases
- 11:33Athena in AWS ecosystem
- 11:56Trino engine behind Athena
- 14:24File formats supported
- 16:59Schema evolution and formats
- 18:27Real-world Athena use case
- 20:54Athena challenges and fixes
- 21:59Query execution and retries
- 23:08ACID and cost model
- 24:34Performance and cost tips
- 25:58Enterprise security in Athena
- 27:10Athena APIs and SDKs
related talks
25mA simple serverless data pipeline
50mServerless SQL Pipelines
39mBig Data Analytics With AWS Athena
49mAnalytics in SQL Server and Azure SQL Database
52mImplement a Serverless Data Analytics strategy
25mData Gone in 60 Seconds: A Serverless ETL Heist
50mSQL 2022 and S3 – Dell’s overview of consuming object data via Microsoft SQL
14mMulticloud Analytics Made Simple