Amazon Redshift
Amazon Redshift is a fast, petabyte-scale, fully managed SQL data warehouse service from Amazon Web Services that lets you analyze data with standard SQL and existing business intelligence tools such as Amazon QuickSight or Tableau, without managing the underlying warehouse infrastructure.
AI Practitioner focus
- Redshift is a managed data warehouse for structured analytics, feature/label exploration, and business-result measurement around AI systems.
- Use Redshift Serverless for variable analytics without cluster management. It is not a substitute for a model endpoint or the primary vector-store examples named in the current exam objectives.
Key points
- Operates as a relational database management system, supporting Online Analytical Processing (OLAP) database workloads.
- Runs complex analytic queries on semi-structured and structured data using query optimization, columnar-based storage, and Massively Parallel Query Execution (MPP).
- Redshift Spectrum queries objects (files) directly on Amazon S3 without loading them into the warehouse.
- Can automatically copy snapshots (automated or manual) of a cluster to another AWS Region.
- Efficiently analyzes petabyte-scale data using existing BI tools and standard SQL, without managing the data warehouse infrastructure (per AWS Terminology.xlsx).
- Redshift Serverless runs and automatically scales the warehouse with no cluster to provision or manage, for intermittent or unpredictable usage.
Pricing
- On-demand pricing charged by the hour, with no commitments and no upfront costs.
- Charged based on the type and number of nodes used in the Redshift cluster.
- Redshift Spectrum is charged based on the number of bytes scanned, rounded up to a 10 MB minimum per query.