Amazon Redshift
by Amazon Web Services
AWS’s cloud data warehouse. It runs in two modes: Redshift Serverless, which provisions and scales capacity automatically,
and provisioned clusters, where the user chooses node types and counts and resizes them manually.
Core model
- Serverless — capacity is provisioned and scaled automatically; the user is billed for compute used
- Provisioned clusters — nodes are added or removed with a cluster resize; clusters can be paused and resumed manually
- Data sources — a warehouse can query Redshift managed storage and an S3 data lake
- Datashares — Serverless can access datashares published from provisioned warehouses
- Reservations — Serverless capacity can be reserved for one or three years for lower compute cost; usage above the reservation is billed on demand
- Regional rollout — Serverless expanded to new AWS regions during 2026, including Asia Pacific (New Zealand), Asia Pacific (Melbourne) and Canada West (Calgary)
Position
Redshift is the warehouse counterpart to Amazon Athena, which queries S3 in place without storage or clusters.
Teams often use both: Redshift for curated warehouse tables and Athena for ad-hoc lake queries. Compare with
BigQuery on Google Cloud.
Related
- Amazon Athena
- BigQuery
- Dosi — compiles semantic models to Redshift SQL