Amazon Redshift
What it is: AWS cloud data warehouse. Columnar storage, massively parallel processing (MPP). SQL-based analytics at petabyte scale.
What It Does Best
AWS integration. Native integration with S3, EMR, Glue, Kinesis. Seamless data movement within AWS ecosystem.
Mature and stable. Decade-old product, battle-tested. Extensive documentation, large community.
Spectrum. Query data in S3 without loading. Extend warehouse to data lake.
Key Features
Columnar storage: PostgreSQL-based with columnar compression
Massively parallel: Distribute queries across many nodes
Redshift Spectrum: Query S3 data directly without loading
Concurrency scaling: Auto-scale for concurrent queries
Materialized views: Precompute query results
Pricing
On-demand: $0.25-5.00/hour depending on node type
Reserved: 50-75% savings with 1-3 year commitments
Serverless: $0.50 per RPU-hour, auto-scaling
Typical cost: $1,000-10,000+/month for most workloads
When to Use It
✅ AWS-centric data infrastructure
✅ Need mature, stable data warehouse
✅ Large data volumes (TB-PB scale)
✅ Team knows PostgreSQL SQL
✅ Query S3 data with Spectrum
When NOT to Use It
❌ Multi-cloud strategy (vendor lock-in)
❌ Small datasets (Postgres cheaper)
❌ Need fastest performance (Snowflake often faster)
❌ Frequent schema changes (tuning required)
❌ Unpredictable workloads (Snowflake better scaling)
Common Use Cases
Business intelligence: Power BI, Tableau, Looker dashboards
Data warehouse: Central repository for analytics
Log analysis: Query application and server logs at scale
Financial reporting: Aggregate transactional data
Data lake queries: SQL on S3 with Spectrum
Redshift vs Alternatives
vs Snowflake: Redshift cheaper on AWS, Snowflake easier to use and faster
vs BigQuery: BigQuery serverless and faster, Redshift more control
vs Athena: Redshift better for frequent queries, Athena for ad-hoc on S3
Unique Strengths
AWS native: Deep integration with AWS services
Spectrum: Query S3 without ETL
PostgreSQL compatibility: Familiar SQL dialect
Mature product: Well-documented, large user base
Bottom line: Solid choice for AWS shops. More operational overhead than Snowflake, but cheaper and proven at scale. Good for teams already deep in AWS ecosystem. Consider Snowflake if multi-cloud or want easier management.