Reviewed by Aditya Kumar · Last reviewed 2026-03-24
**STAR**: S3 (durable, scalable lake), Glue (managed Spark, no cluster ops), Athena (pay-per-query), Lambda (orchestration). **Why**: Serverless—no cluster management, pay-per-use, auto-scale. S3 with lifecycle for cost. Glue bookmarks for incremental. Athena for analyst...
STAR: S3 (durable, scalable lake), Glue (managed Spark, no cluster ops), Athena (pay-per-query), Lambda (orchestration). Why: Serverless—no cluster management, pay-per-use, auto-scale. S3 with lifecycle for cost. Glue bookmarks for incremental. Athena for analyst self-service. Evaluated EMR for heavy Spark; chose Glue for lower ops. Result: 50 pipelines, 100TB/mo, 2 engineers. Best practice: Document decisions; IaC (Terraform) for reproducibility; match service to use case.
Pro-Move: 'We wrote an ADR for Glue vs EMR—Glue for 80% of jobs; EMR for ML and graph; decision based on 2-week POC.' Red Flag: 'We used Glue' without rationale—why Glue over EMR?
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.