Skip to main content
This lesson explains Amazon Redshift: what it is, why you need a data warehouse, Redshift’s architecture and core features, and common use cases for analytics at scale. Why a data warehouse? Data warehouses consolidate data from many sources — databases, log systems, and cloud object stores — to create a unified view of organizational data. That unified view is essential for analytics and business intelligence because it enables cross-functional analysis across historical and current data. Key differences from OLTP databases:
  • OLTP systems are optimized for transactional operations (inserts, updates, deletes).
  • Data warehouses are optimized for analytical, read-heavy workloads with complex, long-running queries and reporting across large datasets.
Common capabilities of data warehouses:
  • Store historical data for trend analysis and comparisons.
  • Include ETL pipelines to clean, transform, and aggregate data before analysis.
  • Operate on structured and aggregated data to simplify analytics.
  • Scale compute and storage independently to accommodate growth.
  • Provide cost-effective read-optimized storage and query processing.
A presentation slide titled "Why do we need Data Warehouse?" with an illustration of people working on a server-like data grid. Colored boxes list benefits like Data Integration, Optimized for Analytics, Historical Analysis, Data Transformation, Structured Data, Scalability, and Cost Efficiency.
What is Amazon Redshift?
  • Amazon Redshift is a fully managed, petabyte-scale cloud data warehouse on AWS.
  • It provides a relational, SQL-based platform optimized for BI, reporting, and analytics.
  • Redshift uses columnar storage, compression, and massively parallel processing (MPP) to deliver fast query performance on very large datasets.
Redshift architecture — main components
  • Cluster: The fundamental deployment unit. A cluster contains one or more nodes.
    • Leader node: Present when there are two or more compute nodes. The leader parses incoming SQL, generates query plans, coordinates compute nodes, and returns final results to clients.
    • Compute nodes: Store data, execute query fragments, and send intermediate results back to the leader for aggregation.
  • Redshift Managed Storage (RMS): Separates durable storage from compute. It uses S3 for petabyte-scale durability while local SSDs on nodes act as a high-performance cache. RA3 nodes exemplify independent compute and storage scaling.
  • Database: Each cluster hosts one or more databases. Redshift implements a SQL-compatible engine influenced by PostgreSQL but has distinct behaviors and optimizations.
Redshift component summary: How the leader and compute nodes work together
  • The leader node parses SQL, builds an execution plan, compiles plan elements, and assigns tasks to compute nodes.
  • Compute nodes execute assigned operations on their data slices, returning intermediate results to the leader for final aggregation.
  • You can scale compute by adding nodes or changing node types; RA3 enables independent storage scaling.
A presentation slide titled "Redshift – Features" showing a grid of colorful rounded rectangles. Each rectangle lists an AWS Redshift capability (e.g., Columnar Storage, MPP, Data Compression, Scalability, Security, Automated Backups, High Availability, Concurrency, Materialized Views, Data Lake Integration, etc.).
Core features and capabilities Typical Redshift use cases
  • Forecasting and predictive analytics (integrate with Amazon SageMaker for ML workflows).
  • Secure data sharing across AWS accounts and external partners without duplicating data.
  • Dashboards and reporting with QuickSight, Tableau, Power BI, and similar tools.
  • Centralized, performant SQL access for multiple client applications and languages.
  • Hybrid analytics combining structured Redshift data with semi-structured or unstructured data in S3.
When to choose Redshift
  • You need a fully managed, SQL-based data warehouse for petabyte-scale analytics.
  • You require fast aggregated queries across large historical datasets.
  • You want native integration with AWS analytics and ML services.
  • You prefer managed scaling and snapshot-based backup/recovery options.
Redshift is based on PostgreSQL and supports many PostgreSQL-compatible features, but it is not a drop-in replacement. Differences exist in concurrency, distribution keys, supported extensions, and certain data types. Test queries and schemas when migrating from PostgreSQL or other data platforms.
Summary
  • Amazon Redshift is a fully managed, petabyte-scale cloud data warehouse optimized for analytics.
  • A leader node coordinates queries while compute nodes store data and run distributed query fragments.
  • Columnar storage, MPP, and automatic compression deliver efficient, high-performance analytics.
  • Built-in security, snapshots, and scaling options support production analytics workloads.
A presentation slide titled "Summary Redshift Provisioned" that states Amazon Redshift is a fully managed, petabyte-scale cloud data warehouse. It also lists node types, describing the Leader Node (query coordination) and Compute Nodes (store data and execute queries).
Operational and security highlights
  • Compression selection and data sampling: Redshift recommends compression encodings and samples data to minimize storage and I/O automatically.
  • Encryption: AES-256 for data at rest and SSL/TLS for data in transit; integrates with AWS KMS for key management.
  • Access controls: IAM integration and fine-grained, column-level access control where applicable.
  • Backups and scaling: Automated/manual snapshots for recovery and the ability to temporarily add capacity to handle peak workloads.
A presentation slide titled "Summary Redshift Provisioned" with a teal left panel. The right side shows three numbered points describing AES-256/SSL encryption, fine-grain column-level access controls, and automatic cluster capacity scaling.
Links and references In this lesson you learned what a data warehouse provides, the architectural elements of Amazon Redshift, its key features, and typical scenarios where Redshift delivers value for analytics workloads.

Watch Video