ACKF logoACKF
Database

Amazon Redshift

View on GitHub

Amazon Redshift

A fully managed, petabyte-scale data warehousing service that provides a columnar, MPP (massively parallel processing) SQL analytic engine. It operates as a managed cluster or serverless endpoint and integrates with data lakes and ETL pipelines to run analytic queries. It normally fits as the central analytical datastore in an AWS analytics architecture for BI, reporting, and large-scale read-optimized queries.

🗂 Resource Category

Analytics • Database

🧠 Exam Memory Hook

Think: "Large-scale SQL analytics + columnar MPP + S3 querying = Amazon Redshift"

📖 Ownership

Classification: AWS Managed Service

AWS responsibilities: AWS operates, maintains, and patches the underlying physical infrastructure and the managed Redshift service platform, including the control plane and managed database runtime; AWS is responsible for platform availability operations and underlying hardware lifecycle activities. AWS secures the infrastructure and the managed service components by applying infrastructure-level patches and platform maintenance according to the service maintenance model. AWS maintains the managed service features such as the cluster control plane, storage layer provided by Redshift, and service APIs.

Customer responsibilities: The customer configures clusters or serverless endpoints, network placement (VPC subnets, security groups), database schemas, user accounts and permissions, queries, workload management settings, data loading and ETL/ELT processes, and backup retention settings. The customer secures data at rest and in transit through encryption settings, configures IAM access, manages database users and roles, monitors query performance and resource usage, and implements application-level protections and auditing. The customer is responsible for optimizing queries, schema design, and data distribution to meet performance and cost goals.

Patching responsibilities: AWS patches physical infrastructure, the hypervisor layer, and the managed Redshift platform and service runtime as part of platform maintenance. Customers do not patch the underlying OS or hypervisor for managed Redshift clusters; however, customers are responsible for patching client-side components such as JDBC/ODBC drivers, client applications, user-defined functions and libraries deployed by the customer, and any application runtimes they control.

🏗 Typical Architecture

💡 Top 5 Features

  • Columnar storage and an MPP query engine optimized for large-scale analytic queries.
  • Redshift Spectrum to run SQL queries directly against data stored in Amazon S3 using the Glue Data Catalog.
  • Workload management and concurrency controls to prioritise and route queries and manage resource queues.
  • Snapshot-based backup and restore capabilities and point-in-time restore when configured.
  • SQL interface with JDBC and ODBC support and PostgreSQL-compatible client connectivity.

✅ Top 5 Use Cases

  • Centralized data warehouse for business intelligence and standardised reporting across enterprise data sources.
  • Querying a data lake using Redshift Spectrum to combine hot warehouse data and colder S3 data without full ingestion.
  • ELT pipelines that load transformed data into Redshift for high-performance analytical queries from ETL tools or COPY jobs.
  • Ad hoc, interactive analytics and dashboarding from BI tools using JDBC/ODBC connections to Redshift.
  • Federated analytics that combine Redshift data with live operational databases using federated query capabilities.

🏗 Architecture Placement

Amazon Redshift typically sits as the analytical datastore in a regional VPC and receives data from ETL/streaming sources and data lakes; ingestion can come from Amazon S3, Kinesis Data Firehose, or direct data-loading jobs and it is commonly queried by BI tools and analytics applications. It integrates with the AWS Glue Data Catalog for schema discovery and with CloudWatch for monitoring. Placement may be cluster-based (provisioned nodes) or serverless depending on configuration.

🎯 Commonly Used With

- Amazon S3
- AWS Glue
- Amazon Kinesis Data Firehose
- Amazon CloudWatch
- AWS Identity and Access Management

🌍 5 Real-World Examples

  • A retail analytics platform consolidates sales, inventory, and clickstream data into Redshift for daily business reporting and supply chain analysis.
  • A financial services team stores normalized trade and market data in Redshift to run compliance reporting and risk aggregation queries.
  • A healthcare analytics environment uses Redshift to run cohort analyses and operational reporting on de-identified patient and claims data.
  • An ad-tech analytics pipeline lands event data in S3, uses Spectrum from Redshift to join with managed datasets, and serves aggregated metrics to dashboards.
  • A telecommunications operations team loads network performance data into Redshift to analyse historical trends and capacity planning.

🎓 AWS Exam Clues

  • Choose Redshift when you need a managed, columnar MPP data warehouse for large-scale SQL analytics.
  • Consider integration with Amazon S3 and Redshift Spectrum when you must query data that remains in a data lake without full ingestion.
  • Evaluate workload management, concurrency, and scaling behaviour for high-concurrency BI workloads and dashboarding.
  • Confirm whether provisioned nodes or serverless Redshift models better match variable workloads and administrative preferences.
  • Account for storage and compute separation options and snapshot retention when designing backup and restore procedures.

📝 Quick Revision

Amazon Redshift is a managed, columnar MPP data warehouse for large-scale SQL analytics and BI workloads. Use it when you need a central analytic store with integrations to S3 and Glue; plan for workload management, query optimisation, security configuration, and snapshot-based backups.

🏷 Keywords

columnar storage • MPP • Redshift Spectrum • COPY command • snapshots • workload management • concurrency scaling • materialized views • federated queries • JDBC/ODBC • Glue Data Catalog • data lake query