Reviewed by Aditya Kumar · Last reviewed 2026-03-24
Architecture: Ingestion via API Gateway + Lambda or Kinesis → S3 landing zone (JSON/Parquet) partitioned dt=YYYY-MM-DD. Glue Crawlers or manual schema → Athena tables. Query: Funnels, sessions, A/B tests—e.g., conversion by landing page. Why S3 + Athena: Decoupled...
Architecture: Ingestion via API Gateway + Lambda or Kinesis → S3 landing zone (JSON/Parquet) partitioned dt=YYYY-MM-DD. Glue Crawlers or manual schema → Athena tables. Query: Funnels, sessions, A/B tests—e.g., conversion by landing page. Why S3 + Athena: Decoupled storage/compute; pay per query; no cluster. Scalability: S3 unlimited; Athena concurrency unlimited. Cost: Partition by date and campaign_id; use Parquet—10x compression, column pruning. Best practice: Raw/curated/aggregated zones; lifecycle policies; query result caching; Redshift/QuickSight for dashboards if needed.
Red Flag: Single-zone, unpartitioned JSON—slow and expensive. Pro-Move: 'We partition by dt and campaign_id, use Parquet—analysts run 50+ Athena queries/day; we cache results and use lifecycle for raw.'
According to DataEngPrep.tech, this is one of the most frequently asked Cloud/Tools interview questions, reported at 1 company. DataEngPrep.tech maintains an editor-reviewed database of 1,863 data engineering interview questions across 7 categories.