Amazon Redshift
Fully managed, petabyte-scale data warehouse for SQL analytics and BI reporting on large datasets.
Verified against AWS documentation · 2026-09-09
Key facts
- 01OLAP, not OLTP — built to scan and aggregate billions of rows, not to serve single-row app lookups
- 02Columnar storage + MPP: a leader node plans the query, compute nodes run it in parallel on their own slice of data
- 03Two flavours: Serverless (pay per RPU-second while active, no cluster to size) and Provisioned (node-hours, pause/resume, Reserved Instances)
- 04Spectrum queries data sitting in S3 directly — no loading, join warehouse tables to lake files in one SQL statement
- 05Reality check: AWS now steers new clusters to Graviton-based RG nodes with a built-in lake query engine; exam material still says RA3 and DC2
Exam triggers
Phrases and scenarios that point here.
- "Data warehouse", "OLAP", "business intelligence", "petabyte-scale analytics", "complex joins across large tables"
- Reporting queries are crushing the production database → move analytics to Redshift, leave RDS to the app
- Analytics run only at month-end or business hours → Redshift Serverless, no capacity planning
- Query existing S3 data from the warehouse without an ETL load → Redshift Spectrum
Not this service
- Occasional ad hoc SQL over S3, nothing to run or manage → Athena
- Transactional app database, single-row reads and writes → RDS or Aurora
- Spark, Hadoop or custom processing instead of SQL → EMR
- The dashboards and charts themselves → QuickSight (Redshift is what it queries)
Use cases
- 01One reporting warehouse fed nightly from orders, CRM and billing, so finance runs a single query instead of reconciling three exports
- 02Five years of clickstream in one place to answer questions no single app database can — which acquisition channel actually produces repeat buyers
- 03BI backend for QuickSight, Tableau or Power BI, where a few hundred analysts hit the same tables every Monday morning
- 04Hot months live in the warehouse, cold years stay in S3 as Parquet, and Spectrum joins both in one query — you keep the history without paying warehouse storage for it
- 05Serverless for a warehouse that only wakes up for the month-end close, so the other 25 days cost nothing in compute
How it's charged
- 01Two different meters: Provisioned bills per node-hour whether or not a query runs; Serverless bills RPU-hours per second, only while active, with a 60-second minimum
- 02Storage is billed separately as Redshift Managed Storage, per GB-month, independent of node count — so you scale data without adding nodes
- 03Spectrum is $5 per TB scanned from S3 (10 MB minimum per query); compressing and using Parquet cuts the bill by orders of magnitude. Serverless and RG include lake queries with no separate Spectrum charge
- 04Concurrency Scaling: one free hour of credits earned per 24 hours of cluster uptime, accruing up to 30 hours — enough for most workloads before the per-second rate kicks in
- 05Reserved Instances (1 or 3 year) discount steady provisioned clusters; Serverless Reservations save up to 45% on a 3-year term. Pausing a cluster stops compute billing but you still pay for snapshots
